{"content": "the await tag does this in one step \u2014 [!await] makes the rest of the turn what you wait for; [!await on=(\"\", \"helper:\")] waits on those; it runs only when it opens the last text of your turn", "meta": {"from": "journal"}} {"content": "the full suite on the invocation funnel and route ranking came back - Run full\u2026", "meta": {"from": "journal"}} {"content": "fact 27 \u2014 Running src/journal.py against the live .journal root starts a\u2026 \u2014 Seen 2026-10-01: with VERSION bumped in the tree, src/journal.py --root .journal saw the live server as another build and started a new one on a new port (8431, 8432), and the user's tabs kept losing connection. Run source builds against a throwaway or demo root (~/projects/demo-crumb/.journal), never the live one, until the release is installed.; rule 57 \u2014 Never merge the overnight refactor into main before its pull request\u2026 \u2014 Messages 15005, 15006, 15109, 15110 (2026-10-04): all refactor work goes on branch overnight-refactor and reaches the user as one pull request, which they read in the morning; nothing of it is merged into main until they say so. Hotfixes the user explicitly asks for go to main at once and are merged into the branch.", "meta": {"from": "journal"}} {"content": "your command ran 32s in the foreground and was moved to the background \u2014 carry on with other work; you are told when it ends. Start a command you expect to take long in the background yourself", "meta": {"from": "journal"}} {"content": "law L3 \u2014 Read narrowly - grep for the line, sed a range, head the file; never\u2026 \u2014 Everything a tool returns stays in the context for good and is paid for on every turn after it. Search before you read, read the range you need, and cap output with grep, head or tail. Read a whole file only when you need all of it.", "meta": {"from": "journal"}} {"content": "work 1 is still open \u2014 end it or park it before you stop: journal work end 1 --how \"\", or journal work park 1 \"\"", "meta": {"from": "journal"}} {"content": "your command ran 34s in the foreground and was moved to the background \u2014 carry on with other work; you are told when it ends. Start a command you expect to take long in the background yourself", "meta": {"from": "journal"}} {"content": "the hook hit an error \u2014 journal: the hook hit an error and kept going; the last of it is below and the whole of it is in .journal/runtime/engine.log. Fix it, then say so. the hook got no answer from the server 1 times (codes 000)", "meta": {"from": "journal"}} {"content": "fact 9 \u2014 Every public method on a controller becomes a journal command \u2014 The CLI is generated from the controllers: each public method of Controller, or of a typed controller, turns into journal . A helper added to the base class therefore becomes a command on every type \u2014 which is how journal handled and journal refuse came to exist, from the CRUD funnel and the refusal funnel. An internal helper on a controller is named with a leading underscore, as _shaped and _status already are, or it ships as a command nobody meant.", "meta": {"from": "journal"}} {"content": "rule 27 \u2014 Name a declaration with the word a reader already knows \u2014 An attribute, a variable or a field gets the ordinary programming word for what it holds, not an evocative one. was, heard and alone were poetry; aliases, notify_actions and urgent_actions are what they are. The test: could a reader who has never seen this codebase guess what it holds from the name alone? Prose belongs in the help text and the abstract, where it is read as prose. This does not license abbreviations \u2014 a plain word in full, not a short one.; rule 55 \u2014 Always dispatch Codex helpers on gpt-6-sol \u2014 The user's word, message 13431: switch the codex agents to GPT-6-Sol and make it their default. ~/.codex/config.toml names it as the default model too.; rule 56 \u2014 Helpers are for work that writes; subagents read, research and design \u2014 The user, message 13464: there must be a clear distinction. A subagent can be dispatched for anything read-only: research, review, design. A helper is for actual work that writes, best in its own worktree when the work is separate. Dieter designing in Claude Design should have been a subagent, not a helper.", "meta": {"from": "journal"}} {"content": "law L5 \u2014 Every subagent dispatch names the agent - a human name, a little\u2026 \u2014 A name is how the user and the chat tell subagents apart and how they are messaged later; an id or a task line is not a name. Start the dispatch's description with the name, a colon, then the task, such as \"Dr. Einstein: profile the slow hooks\" or \"Coco Rams: draw the plan card\". A designer can borrow from famous designers, a researcher from famous scientists, mixed up for fun.", "meta": {"from": "journal"}} {"content": "work 1 is still open \u2014 end it or park it before you stop: journal work end 1 --how \"\", or journal work park 1 \"\"", "meta": {"from": "journal"}} {"content": "the damaged-row boot test rerun without the leaked JOURNAL_ENV came back\u2026", "meta": {"from": "journal"}} {"content": "rule 41 \u2014 Keep moving, run the whole suite before every commit, never wait \u2014 The full suite runs in about seven seconds: .venv/bin/python -m pytest -q --timeout=300 -n auto. Run it before every commit instead of picking tests by name. Group rows that sit in the same code into one sitting: write them all, test once, commit once. And never wait, not for a subagent, a build, or an answer you can carry on without. Dispatch it and keep working. If you truly are waiting on something, say so in the work log.; rule 46 \u2014 Only commit and push once the whole journal is proven to boot \u2014 Messages 2178 and 2179. A release that has not been started for real can crash every project that installs it, as 2.78.3 did for Codex and 2.84.0 did to project records. Before every commit and push: the full suite passes, including tests/test_it_boots.py, which installs a packed copy into a fresh project, launches Claude and Codex from it, writes a project record, upgrades again and checks the record survives. When a change touches launching, installing or upgrading, also start a real journal in a scratch project and close it properly afterwards, leaving no process behind.; rule 50 \u2014 Everything the user does is doable in the viewer \u2014 Message 6710 (2026-09-23): the user never uses the CLI, only the UI; everything should be doable from the viewer. The journal commands are for agents; any action meant for the user (making boards, confirming, accepting, hosting, watching an agent) needs its place in the viewer.", "meta": {"from": "journal"}} {"content": "fact 23 \u2014 Every upgrade brings system sequences and their triggers in line\u2026 \u2014 install.py runs ship_sequences after the migrations on each upgrade, so features/sequences/shipped.py is the whole source: change its wording and the next upgrade updates every journal, no migration needed. Shipped rows carry system=True and are read-only for everyone but SYSTEM (controllers/base.py _shipped).", "meta": {"from": "journal"}} {"content": "the log tag does this in one step \u2014 [!log:N] makes the turn itself the log entry; it runs only when it opens the last text of your turn", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 the files changed since the last check (`stored.py`, `recor\u2026 \u2014 Code Commandments \u2014 the files changed since the last check (`stored.py`, `record.py`) breaks a rule. Fix it now, at its SOURCE, while the code is still in front of you: \u00b7 \u2022 python-mutable-value-object at /Users/jessegall/projects/agent-journal/.claude/worktrees/main-edsger-lockwood/src/engine/stored.py:72 \u00b7 LOAD the skill `commandments-python-value-objects` before fixing \u2014 load it even if you believe you already have. \u00b7 \u2022 python-flag-argument at /Users/jessegall/projects/agent-journal/.claude/worktrees/main-edsger-lockwood/src/engine/record.py:115 \u00b7 LOAD the skill `commandments-python-behaviour-per-method` before fixing \u2014 load it even if you believe you already have. \u00b7 Run `commandments info ` if a rule is not one you recognise. This check reads a file at a time, so it is not the whole picture \u2014 `judge` still is.", "meta": {"from": "journal"}} {"content": "rule 54 \u2014 Settings and feature switches are read at boot and on change, never\u2026 \u2014 The user, message 13349: the application boots, determines every feature and setting once, and re-evaluates only when something changes, such as a setting or a plugin. Never lazy-load settings.", "meta": {"from": "journal"}} {"content": "fact 24 \u2014 An answer followed by tool calls can be missing from Claude's\u2026 \u2014 Seen 2026-09-24 for messages 9391-9404: text blocks opening with [!reply:n] that were followed by tool calls never appeared in the session's jsonl (only thinking and tool_use rows did), so the journal never saw them and the replies were lost. When a turn goes on after answering, send the answer with journal message reply \"\" instead of the tag.", "meta": {"from": "journal"}} {"content": "fact 20 \u2014 A slim supervisor holds the agent and a worker reloads on every build \u2014 Since 2.118.0 (2026-09-23). src/supervisor.py is standard library only and never reloads: journal claude hands its process over to it (os.execv), and it owns the pty and the agent process, relays the terminal, writes the printed and screen captures, listens on the typist socket, restarts the agent in the same session from a relaunch command written to its runtime folder while a restart is pending, and stops it with escalation while draining the pty (an agent cannot finish exiting on macOS while its output is unread). It starts the worker (src/worker.py, which runs runner/worker.py; engine/worker.py stays as an alias for supervisors started before 2.201) and starts it again whenever it exits: RELOAD on a new build, RELAUNCH to restart the agent, STOP to end, HEAL or a quick crash to roll back a build through journal heal. The worker holds everything else: seating the session, the start-up confirm typed through the typist, services, viewer, update check, check-in, and the one-time relaunch of sessions launched before agents/terminal.py LAUNCH. agents/terminal.py holds only journal-side helpers. The server (serve.py) still runs the engines and re-execs itself on a .py change. When the agent exits, the supervisor runs journal ended, which puts set-aside hooks back and stops the server when no session is left.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 28 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 28 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.; work 1 is still open \u2014 end it or park it before you stop: journal work end 1 --how \"\", or journal work park 1 \"\"", "meta": {"from": "journal"}} {"content": "targeted tests on the engine batch (2628, 2629, 2607, 2612) came back - Wait\u2026", "meta": {"from": "journal"}} {"content": "full suite on the engine batch came back - Run full suite for the engine batch\u2026", "meta": {"from": "journal"}} {"content": "rule 35 \u2014 Write clean code - one funnel per kind of operation, never the same\u2026 \u2014 Every kind of operation has one funnel: one method that creates, one that saves, one that refuses, one that formats. A second method that does the same thing under another name splits the behaviour, and the two drift apart. Before writing a method, search for the one that already does it and extend that. scripts/checks/funnels.py finds bodies written twice.", "meta": {"from": "journal"}} {"content": "rule 37 \u2014 Close every to-do explicitly with todo done or a Journal commit\u2026 \u2014 Ending work does not close its row. A to-do is closed by journal todo done --how, or by a commit whose message carries Journal: todos done at column 0, several numbers separated by commas. A row left open after its work landed misleads the next session and auto mode.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you commit \u2014 you've changed 28 judged files since th\u2026 \u2014 Code Commandments \u2014 before you commit: you've changed 28 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "chat etiquette - a line from the journal is an instruction, not a message\u2026 \u2014 a turn that only handles a journal line needs no words: act on it, or say once in the chat what you wait on, then carry on; what the user needs to know still goes to the chat", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you commit \u2014 you've changed 27 judged files since th\u2026 \u2014 Code Commandments \u2014 before you commit: you've changed 27 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.; rule 44 \u2014 Release a new version after every significant change \u2014 The user, message 13219 (2026-10-01): 'Don't forget to release new versions every time you do something significant.' Bump VERSION, add a CHANGELOG entry, push main and the tag. Replaces message 5929's release-on-request.; rule 51 \u2014 Every finished feature is committed, pushed and released with a new\u2026 \u2014 Message 9207 (2026-09-24): when a new feature is ready, commit, push and publish a new tag. This is the user's standing word for releasing, so rule 44's only-when-the-user-says is met by it for finished features; fixes in between wait for the next feature or a patch the user asks for.", "meta": {"from": "journal"}} {"content": "12 facts standing, read them \u2014 9. Every public method on a controller becomes a journal command; 13. This live session runs the installed copy in .journal/journal.pyz; 18. cProfile inflates the slow-request profiles about tenfold; 20. A slim supervisor holds the agent and a worker reloads on every build; 23. Every upgrade brings system sequences and their triggers in line with the code; 24. An answer followed by tool calls can be missing from Claude's transcript; 25. A designer's install packs the whole tree, half-done server edits included; 26. Claude Code reads agent profiles when a session starts; 27. Running src/journal.py against the live .journal root starts a second server; 28. The designer agent type exists, so design work goes to a subagent; 29. Codex's transcript records the end of every exec session, polled or not; 30. The tunler server refuses TLS for any subdomain without a tunnel; 34 rules in force, read them \u2014 7. journal disable must only ever be run because the user explicitly asked for it,; 9. Research dispatched to a subagent ends in a REPORT for the user, compiled by the; 13. A title names the thing in at most 80 characters and never explains it with a co; 14. A small journal capability is a feature under features, with at most one test; 17. Do not restart the viewer for frontend-only changes; 18. Provider-specific code belongs in providers, never features; 19. Providers report facts; features decide behavior; 22. File distinct user work requests immediately; 27. Name a declaration with the word a reader already knows; 30. The viewer has one API client, and every piece of it does one job; 31. Every finding a reviewing agent reports becomes its own to-do; 35. Write clean code - one funnel per kind of operation, never the same method twice; 36. Clean, DRY, idiomatic before it is committed, never after it is complained about; 37. Close every to-do explicitly with todo done or a Journal commit trailer; 38. Never change the git branch until the user says so, by name; 39. Use only registered exclamation response tags; 40. A feature is named for what it is, never for its machinery; 41. Keep moving, run the whole suite before every commit, never wait; 42. Every user-facing text passes the formatters before it leaves the server; 43. A request or hook over its budget is fixed before the next release; 44. Release a new version after every significant change; 45. No prose words as names in code - said, says, heard, spoke, told, shown, became; 46. Only commit and push once the whole journal is proven to boot; 47. The journal sets itself up once, when the server starts, never per command; 48. The viewer is built from its component library, and pages only compose it; 49. A dialog whose content grows keeps one fixed height, and its content scrolls; 50. Everything the user does is doable in the viewer; 51. Every finished feature is committed, pushed and released with a new version tag; 52. A chat mark for something the user did sits on the user's side; 53. Every new design gets one critique round, and the designer decides; 54. Settings and feature switches are read at boot and on change, never per call; 55. Always dispatch Codex helpers on gpt-6-sol; 56. Helpers are for work that writes; subagents read, research and design; 57. Never merge the overnight refactor into main before its pull request is approved", "meta": {"from": "journal"}} {"content": "rule 42 \u2014 Every user-facing text passes the formatters before it leaves the\u2026 \u2014 Not only a brief. A title, an abstract, an outcome and every section body are read by a person, so each goes through the same formatters on its way to the viewer \u2014 chat turns, activity items, to-do rows, inspector pages, docs alike. One field formatted out of five is not a rule, it is an accident, and it is how a raw tag ended up in the activity list after the tags feature had been stripping them for weeks. When a new field carries words a person reads, it joins the list in the same place.", "meta": {"from": "journal"}} {"content": "that Bash call returned 38,113 characters, the largest this session \u2014 It stays in the context for good. If you were looking for one thing in it, the next read can be narrower: grep for the line, sed a range, head the file.", "meta": {"from": "journal"}} {"content": "rule 47 \u2014 The journal sets itself up once, when the server starts, never per\u2026 \u2014 Messages 2220 and 2224. Discovering features and their handlers, seating the feature rows and the rename sweep happen once, at server boot, and again only when a feature is switched on or off, a plugin changes or an environment is added: features.load keeps a set-up generation per journal (SEATED) and redoes the work only when that generation moves. A command, a request or a hook uses what is already there; nothing in their path may rediscover handlers or rescan folders. A cost that repeats per call is a bug to fix, not a budget to raise.", "meta": {"from": "journal"}} {"content": "fact 13 \u2014 This live session runs the installed copy in .journal/journal.pyz \u2014 The running journal (server, hooks, CLI) runs from .journal/journal.pyz with its viewer and skills in .journal/src, never from the repo. A change in the repo reaches it only through python3 src/journal.py --root .journal upgrade, which packs the zip again. A commit alone changes nothing that is running.", "meta": {"from": "journal"}} {"content": "fact 25 \u2014 A designer's install packs the whole tree, half-done server edits\u2026 \u2014 2026-09-25: Eames and Saul run python3 src/journal.py --root .journal upgrade after their viewer builds; it packs every file in src, so a server handler I was halfway through writing went live and raised on every PostToolUse hook. While designers work in parallel, keep server edits whole between tool calls (write and test in the scratchpad first), and reinstall after reverting anything.", "meta": {"from": "journal"}} {"content": "Four helpers were stopped at 00:55 and again resumed; until the cause is found: never run pkill or killall with a pattern; stop only processes you started yourself, by their exact pid.", "meta": {"from": "main"}} {"content": "Code Commandments \u2014 the files changed since the last check (`http.py`, `viewer.\u2026 \u2014 Code Commandments \u2014 the files changed since the last check (`http.py`, `viewer.py`, `git_view.py`, `project_files.py`) breaks a rule. Fix it now, at its SOURCE, while the code is still in front of you: \u00b7 \u2022 python-dict-return-bag at /Users/jessegall/projects/agent-journal/.claude/worktrees/main-edsger-lockwood/src/engine/git_view.py:22; python-dict-return-bag at /Users/jessegall/projects/agent-journal/.claude/worktrees/main-edsger-lockwood/src/engine/git_view.py:33 \u00b7 LOAD the skill `commandments-python-value-objects` before fixing \u2014 load it even if you believe you already have. \u00b7 Run `commandments info ` if a rule is not one you recognise. This check reads a file at a time, so it is not the whole picture \u2014 `judge` still is.", "meta": {"from": "journal"}} {"content": "rule 40 \u2014 A feature is named for what it is, never for its machinery \u2014 Messages 599, 600 and 703. A feature is a capability the user would name and would think of switching off. File tracking, a write gate, a phrase bank, a tree diff are services used inside a feature, not features of their own: they live in the feature they serve. Before adding a directory under features/, say what the user would call it; if the answer names a mechanism, it belongs inside something else. Report 16 holds the grouping this implies.", "meta": {"from": "journal"}} {"content": "rule 48 \u2014 The viewer is built from its component library, and pages only\u2026 \u2014 Message 4258. Every visual piece the viewer shows more than once, or that a user would recognise as the same kind of thing (a dialog, a side panel or inspector, a dropdown, a list row, a switch, a button), is one component in web/src/kit, extracted aggressively, and every page composes those components instead of building its own copy. Before writing markup or styles in a page, look for the kit component that already does it and extend it with a prop; a second hand-built version is a bug. The side panel that animated in but not out, while a separate skill panel did both, is the example.", "meta": {"from": "journal"}} {"content": "your command ran 31s in the foreground and was moved to the background \u2014 carry on with other work; you are told when it ends. Start a command you expect to take long in the background yourself", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 19 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 19 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite on the B3 and B4 moves came back - Wait for full suite result\u2026", "meta": {"from": "journal"}} {"content": "fact 9 \u2014 Every public method on a controller becomes a journal command \u2014 The CLI is generated from the controllers: each public method of Controller, or of a typed controller, turns into journal . A helper added to the base class therefore becomes a command on every type \u2014 which is how journal handled and journal refuse came to exist, from the CRUD funnel and the refusal funnel. An internal helper on a controller is named with a leading underscore, as _shaped and _status already are, or it ships as a command nobody meant.; rule 36 \u2014 Clean, DRY, idiomatic before it is committed, never after it is\u2026 \u2014 The user should never be the one who finds duplication, dead code, a clumsy name or a pattern the codebase does not use. Read the diff before every commit as a reviewer would, and fix what is not clean then, not in a follow-up after a complaint.; rule 41 \u2014 Keep moving, run the whole suite before every commit, never wait \u2014 The full suite runs in about seven seconds: .venv/bin/python -m pytest -q --timeout=300 -n auto. Run it before every commit instead of picking tests by name. Group rows that sit in the same code into one sitting: write them all, test once, commit once. And never wait, not for a subagent, a build, or an answer you can carry on without. Dispatch it and keep working. If you truly are waiting on something, say so in the work log.; rule 46 \u2014 Only commit and push once the whole journal is proven to boot \u2014 Messages 2178 and 2179. A release that has not been started for real can crash every project that installs it, as 2.78.3 did for Codex and 2.84.0 did to project records. Before every commit and push: the full suite passes, including tests/test_it_boots.py, which installs a packed copy into a fresh project, launches Claude and Codex from it, writes a project record, upgrades again and checks the record survives. When a change touches launching, installing or upgrading, also start a real journal in a scratch project and close it properly afterwards, leaving no process behind.; rule 50 \u2014 Everything the user does is doable in the viewer \u2014 Message 6710 (2026-09-23): the user never uses the CLI, only the UI; everything should be doable from the viewer. The journal commands are for agents; any action meant for the user (making boards, confirming, accepting, hosting, watching an agent) needs its place in the viewer.", "meta": {"from": "journal"}} {"content": "rule 57 \u2014 Never merge the overnight refactor into main before its pull request\u2026 \u2014 Messages 15005, 15006, 15109, 15110 (2026-10-04): all refactor work goes on branch overnight-refactor and reaches the user as one pull request, which they read in the morning; nothing of it is merged into main until they say so. Hotfixes the user explicitly asks for go to main at once and are merged into the branch.", "meta": {"from": "journal"}} {"content": "that Bash call returned 45,443 characters, the largest this session \u2014 It stays in the context for good. If you were looking for one thing in it, the next read can be narrower: grep for the line, sed a range, head the file.", "meta": {"from": "journal"}} {"content": "rule 38 \u2014 Never change the git branch until the user says so, by name \u2014 The work happens on the branch the user named. That was main until message 5929 and question 80 (2026-09-23), which moved the sins work to the branch sins. Do not create, switch to or merge any other branch unless the user names it in their own words.", "meta": {"from": "journal"}} {"content": "fact 18 \u2014 cProfile inflates the slow-request profiles about tenfold \u2014 The faults feature writes a profile when a request passes its budget, and the profile is taken with cProfile, which adds per-call overhead. On 2026-09-22 /api/summary profiled at 58ms with 48ms inside Resource.fork's deep copy; with the profiler off the same call ran in 2 to 7ms. Read the profile for where the time goes in relative terms, then time the call with curl before changing anything.", "meta": {"from": "journal"}} {"content": "fact 26 \u2014 Claude Code reads agent profiles when a session starts \u2014 Seen 2026-09-26: after the board-filler's profile in .claude/agents gained its steps and the Grep rule, dispatches from the running session still used the old profile (a 3.5-minute first question, shell grep refused); after the session restarted, the same request took 21 seconds with 4 calls. A change to an agent type reaches only sessions started after it is written.; rule 43 \u2014 A request or hook over its budget is fixed before the next release \u2014 Comment 1151 on this rule. When the faults feature reports a request, a hook or a command slower than its budget, file it as a to-do at once. It does not jump ahead of the work in hand, but no version is published while one is still open: profile it, fix it, and verify the new time before the release goes out. The budget is 50ms, because everything runs locally against files.", "meta": {"from": "journal"}} {"content": "law L5 \u2014 Every subagent dispatch names the agent - a human name, a little\u2026 \u2014 A name is how the user and the chat tell subagents apart and how they are messaged later; an id or a task line is not a name. Start the dispatch's description with the name, a colon, then the task, such as \"Dr. Einstein: profile the slow hooks\" or \"Coco Rams: draw the plan card\". A designer can borrow from famous designers, a researcher from famous scientists, mixed up for fun.", "meta": {"from": "journal"}} {"content": "rule 56 \u2014 Helpers are for work that writes; subagents read, research and design \u2014 The user, message 13464: there must be a clear distinction. A subagent can be dispatched for anything read-only: research, review, design. A helper is for actual work that writes, best in its own worktree when the work is separate. Dieter designing in Claude Design should have been a subagent, not a helper.", "meta": {"from": "journal"}} {"content": "the await tag does this in one step \u2014 [!await] makes the rest of the turn what you wait for; [!await on=(\"\", \"helper:\")] waits on those; it runs only when it opens the last text of your turn", "meta": {"from": "journal"}} {"content": "full suite on the rebased branch came back - Run full suite on the rebased\u2026; rule 35 \u2014 Write clean code - one funnel per kind of operation, never the same\u2026 \u2014 Every kind of operation has one funnel: one method that creates, one that saves, one that refuses, one that formats. A second method that does the same thing under another name splits the behaviour, and the two drift apart. Before writing a method, search for the one that already does it and extend that. scripts/checks/funnels.py finds bodies written twice.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 1 judged file since the\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 1 judged file since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "your chat talked about the journal's workings - \"Still waiting\" \u2014 the user sees replies, reactions, pills and reads themselves; say what the work is instead", "meta": {"from": "journal"}} {"content": "race stress loop and full suite after the note-order fix came back - Stress\u2026", "meta": {"from": "journal"}} {"content": "rule 54 \u2014 Settings and feature switches are read at boot and on change, never\u2026 \u2014 The user, message 13349: the application boots, determines every feature and setting once, and re-evaluates only when something changes, such as a setting or a plugin. Never lazy-load settings.", "meta": {"from": "journal"}} {"content": "full suite on the branch rebased onto the newest overnight-refactor came back\u2026", "meta": {"from": "journal"}} {"content": "rule 37 \u2014 Close every to-do explicitly with todo done or a Journal commit\u2026 \u2014 Ending work does not close its row. A to-do is closed by journal todo done --how, or by a commit whose message carries Journal: todos done at column 0, several numbers separated by commas. A row left open after its work landed misleads the next session and auto mode.", "meta": {"from": "journal"}} {"content": "fact 13 \u2014 This live session runs the installed copy in .journal/journal.pyz \u2014 The running journal (server, hooks, CLI) runs from .journal/journal.pyz with its viewer and skills in .journal/src, never from the repo. A change in the repo reaches it only through python3 src/journal.py --root .journal upgrade, which packs the zip again. A commit alone changes nothing that is running.; fact 20 \u2014 A slim supervisor holds the agent and a worker reloads on every build \u2014 Since 2.118.0 (2026-09-23). src/supervisor.py is standard library only and never reloads: journal claude hands its process over to it (os.execv), and it owns the pty and the agent process, relays the terminal, writes the printed and screen captures, listens on the typist socket, restarts the agent in the same session from a relaunch command written to its runtime folder while a restart is pending, and stops it with escalation while draining the pty (an agent cannot finish exiting on macOS while its output is unread). It starts the worker (src/worker.py, which runs runner/worker.py; engine/worker.py stays as an alias for supervisors started before 2.201) and starts it again whenever it exits: RELOAD on a new build, RELAUNCH to restart the agent, STOP to end, HEAL or a quick crash to roll back a build through journal heal. The worker holds everything else: seating the session, the start-up confirm typed through the typist, services, viewer, update check, check-in, and the one-time relaunch of sessions launched before agents/terminal.py LAUNCH. agents/terminal.py holds only journal-side helpers. The server (serve.py) still runs the engines and re-execs itself on a .py change. When the agent exits, the supervisor runs journal ended, which puts set-aside hooks back and stops the server when no session is left.; fact 23 \u2014 Every upgrade brings system sequences and their triggers in line\u2026 \u2014 install.py runs ship_sequences after the migrations on each upgrade, so features/sequences/shipped.py is the whole source: change its wording and the next upgrade updates every journal, no migration needed. Shipped rows carry system=True and are read-only for everyone but SYSTEM (controllers/base.py _shipped).; fact 24 \u2014 An answer followed by tool calls can be missing from Claude's\u2026 \u2014 Seen 2026-09-24 for messages 9391-9404: text blocks opening with [!reply:n] that were followed by tool calls never appeared in the session's jsonl (only thinking and tool_use rows did), so the journal never saw them and the replies were lost. When a turn goes on after answering, send the answer with journal message reply \"\" instead of the tag.; rule 27 \u2014 Name a declaration with the word a reader already knows \u2014 An attribute, a variable or a field gets the ordinary programming word for what it holds, not an evocative one. was, heard and alone were poetry; aliases, notify_actions and urgent_actions are what they are. The test: could a reader who has never seen this codebase guess what it holds from the name alone? Prose belongs in the help text and the abstract, where it is read as prose. This does not license abbreviations \u2014 a plain word in full, not a short one.; rule 44 \u2014 Release a new version after every significant change \u2014 The user, message 13219 (2026-10-01): 'Don't forget to release new versions every time you do something significant.' Bump VERSION, add a CHANGELOG entry, push main and the tag. Replaces message 5929's release-on-request.; rule 45 \u2014 No prose words as names in code - said, says, heard, spoke, told\u2026 \u2014 Messages 1360 and 1698. The user has said more than once that code must not read like prose: a variable, attribute, property or function is named for what it holds or does (text, command, labels, lines), never with a verb from a story. 'says' on the Design type (1360) and 'said = call.said.lower()' in features/recital.py (1698) are the examples. Rule 27 states the naming rule; this one carries the words, so writing one of them whispers it. Before writing a name, ask whether a reader who has never seen the code would know what it holds.; rule 55 \u2014 Always dispatch Codex helpers on gpt-6-sol \u2014 The user's word, message 13431: switch the codex agents to GPT-6-Sol and make it their default. ~/.codex/config.toml names it as the default model too.", "meta": {"from": "journal"}} {"content": "Thank you, and for the JOURNAL_ENV trap (fixed in tests/isolation.py on the branch). Review of your ten commits: one blocker and three smaller points, to-dos 2656,2657,2658,2659 (journal todo read ): (1) BLOCKER: creating a work, question or environment over HTTP now answers 400 because invoked resolves 'create' through Controller.method, which refuses an aliased word; resolve the method name and add an HTTP create case per aliased type; (2) announce each deferred event in its own try, re-raise the first failure after; (3) spread only lists/tuples and copy inside; (4) refuse environment names that a global /api route would shadow, plus the idiom points in it. Rebase onto overnight-refactor first (it moved 26 commits), commit, report.", "meta": {"from": "main"}} {"content": "rule 42 \u2014 Every user-facing text passes the formatters before it leaves the\u2026 \u2014 Not only a brief. A title, an abstract, an outcome and every section body are read by a person, so each goes through the same formatters on its way to the viewer \u2014 chat turns, activity items, to-do rows, inspector pages, docs alike. One field formatted out of five is not a rule, it is an accident, and it is how a raw tag ended up in the activity list after the tags feature had been stripping them for weeks. When a new field carries words a person reads, it joins the list in the same place.; rule 47 \u2014 The journal sets itself up once, when the server starts, never per\u2026 \u2014 Messages 2220 and 2224. Discovering features and their handlers, seating the feature rows and the rename sweep happen once, at server boot, and again only when a feature is switched on or off, a plugin changes or an environment is added: features.load keeps a set-up generation per journal (SEATED) and redoes the work only when that generation moves. A command, a request or a hook uses what is already there; nothing in their path may rediscover handlers or rescan folders. A cost that repeats per call is a bug to fix, not a budget to raise.", "meta": {"from": "journal"}} {"content": "rule 40 \u2014 A feature is named for what it is, never for its machinery \u2014 Messages 599, 600 and 703. A feature is a capability the user would name and would think of switching off. File tracking, a write gate, a phrase bank, a tree diff are services used inside a feature, not features of their own: they live in the feature they serve. Before adding a directory under features/, say what the user would call it; if the answer names a mechanism, it belongs inside something else. Report 16 holds the grouping this implies.", "meta": {"from": "journal"}} {"content": "the log tag does this in one step \u2014 [!log:N] makes the turn itself the log entry; it runs only when it opens the last text of your turn", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 11 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 11 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for the review fixes came back - Full suite for the review fixes\u2026", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you commit \u2014 you've changed 9 judged files since the\u2026 \u2014 Code Commandments \u2014 before you commit: you've changed 9 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite after rebasing the review fixes came back - Scan and run full suite\u2026", "meta": {"from": "journal"}} {"content": "Taken: your fourteen commits are on overnight-refactor (52bcd162). Next for you, same worktree: the items that waited on you - to-do 2653 is Barbara's (she does it once you are done with src/commands), so for you: Edsger's B3.8 routes module split is still off (feature tests import from commands.http), so instead take to-do 2602 (actions exposed by an opt-in @action marker; prove the marked set equals today's actions()), then 2608 (features reach controllers by class, never by name: BoundJournal.__getattr__/NAMED go). These touch every feature file lightly; the feature helpers have landed or are landing, so rebase often. Commit per to-do, report.", "meta": {"from": "main"}} {"content": "law L3 \u2014 Read narrowly - grep for the line, sed a range, head the file; never\u2026 \u2014 Everything a tool returns stays in the context for good and is paid for on every turn after it. Search before you read, read the range you need, and cap output with grep, head or tail. Read a whole file only when you need all of it.", "meta": {"from": "journal"}} {"content": "fact 9 \u2014 Every public method on a controller becomes a journal command \u2014 The CLI is generated from the controllers: each public method of Controller, or of a typed controller, turns into journal . A helper added to the base class therefore becomes a command on every type \u2014 which is how journal handled and journal refuse came to exist, from the CRUD funnel and the refusal funnel. An internal helper on a controller is named with a leading underscore, as _shaped and _status already are, or it ships as a command nobody meant.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 the files changed since the last check (`controller.py`, `c\u2026 \u2014 Code Commandments \u2014 the files changed since the last check (`controller.py`, `controller.py`, `controller.py`, `controller.py`, `controller.py`) breaks a rule. Fix it now, at its SOURCE, while the code is still in front of you: \u00b7 \u2022 python-blank-string-default at /Users/jessegall/projects/agent-journal/.claude/worktrees/main-edsger-lockwood/src/features/tickets/controller.py:685; python-blank-string-default at /Users/jessegall/projects/agent-journal/.claude/worktrees/main-edsger-lockwood/src/features/journal_laws/controller.py:19; python-blank-string-default at /Users/jessegall/projects/agent-journal/.claude/worktrees/main-edsger-lockwood/src/features/sharing/controller.py:116 \u00b7 LOAD the skill `commandments-python-absence` before fixing \u2014 load it even if you believe you already have. \u00b7 Run `commandments info ` if a rule is not one you recognise. This check reads a file at a time, so it is not the whole picture \u2014 `judge` still is.", "meta": {"from": "journal"}} {"content": "base.py names 5 files in the project \u2014 write the path so the chat can link it: src/controllers/base.py, src/engine/events/base.py, src/features/base.py, src/providers/base.py, src/resources/base.py", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 the files changed since the last check (`controller.py`, `c\u2026 \u2014 Code Commandments \u2014 the files changed since the last check (`controller.py`, `controller.py`, `controller.py`, `controller.py`, `controller.py`) breaks a rule. Fix it now, at its SOURCE, while the code is still in front of you: \u00b7 \u2022 python-blank-string-default at /Users/jessegall/projects/agent-journal/.claude/worktrees/main-edsger-lockwood/src/features/tickets/controller.py:681 \u00b7 LOAD the skill `commandments-python-absence` before fixing \u2014 load it even if you believe you already have. \u00b7 Run `commandments info ` if a rule is not one you recognise. This check reads a file at a time, so it is not the whole picture \u2014 `judge` still is.", "meta": {"from": "journal"}} {"content": "rule 41 \u2014 Keep moving, run the whole suite before every commit, never wait \u2014 The full suite runs in about seven seconds: .venv/bin/python -m pytest -q --timeout=300 -n auto. Run it before every commit instead of picking tests by name. Group rows that sit in the same code into one sitting: write them all, test once, commit once. And never wait, not for a subagent, a build, or an answer you can carry on without. Dispatch it and keep working. If you truly are waiting on something, say so in the work log.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 the files changed since the last check (`controller.py`, `c\u2026 \u2014 Code Commandments \u2014 the files changed since the last check (`controller.py`, `controller.py`, `controller.py`, `controller.py`, `todos.py`) breaks a rule. Fix it now, at its SOURCE, while the code is still in front of you: \u00b7 \u2022 python-blank-string-default at /Users/jessegall/projects/agent-journal/.claude/worktrees/main-edsger-lockwood/src/controllers/todos.py:23 \u00b7 LOAD the skill `commandments-python-absence` before fixing \u2014 load it even if you believe you already have. \u00b7 Run `commandments info ` if a rule is not one you recognise. This check reads a file at a time, so it is not the whole picture \u2014 `judge` still is.", "meta": {"from": "journal"}} {"content": "rule 36 \u2014 Clean, DRY, idiomatic before it is committed, never after it is\u2026 \u2014 The user should never be the one who finds duplication, dead code, a clumsy name or a pattern the codebase does not use. Read the diff before every commit as a reviewer would, and fix what is not clean then, not in a follow-up after a complaint.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 40 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 40 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for the opt-in action marker came back - Wait for the 2602 suite\u2026", "meta": {"from": "journal"}} {"content": "rule 46 \u2014 Only commit and push once the whole journal is proven to boot \u2014 Messages 2178 and 2179. A release that has not been started for real can crash every project that installs it, as 2.78.3 did for Codex and 2.84.0 did to project records. Before every commit and push: the full suite passes, including tests/test_it_boots.py, which installs a packed copy into a fresh project, launches Claude and Codex from it, writes a project record, upgrades again and checks the record survives. When a change touches launching, installing or upgrading, also start a real journal in a scratch project and close it properly afterwards, leaving no process behind.; rule 50 \u2014 Everything the user does is doable in the viewer \u2014 Message 6710 (2026-09-23): the user never uses the CLI, only the UI; everything should be doable from the viewer. The journal commands are for agents; any action meant for the user (making boards, confirming, accepting, hosting, watching an agent) needs its place in the viewer.; rule 57 \u2014 Never merge the overnight refactor into main before its pull request\u2026 \u2014 Messages 15005, 15006, 15109, 15110 (2026-10-04): all refactor work goes on branch overnight-refactor and reaches the user as one pull request, which they read in the morning; nothing of it is merged into main until they say so. Hotfixes the user explicitly asks for go to main at once and are merged into the branch.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 50 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 50 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for controllers by class came back - Full suite for 2608 completed\u2026", "meta": {"from": "journal"}} {"content": "your command ran 34s in the foreground and was moved to the background \u2014 carry on with other work; you are told when it ends. Start a command you expect to take long in the background yourself", "meta": {"from": "journal"}} {"content": "work 3 is still open \u2014 end it or park it before you stop: journal work end 3 --how \"\", or journal work park 3 \"\"", "meta": {"from": "journal"}} {"content": "rule 54 \u2014 Settings and feature switches are read at boot and on change, never\u2026 \u2014 The user, message 13349: the application boots, determines every feature and setting once, and re-evaluates only when something changes, such as a setting or a plugin. Never lazy-load settings.", "meta": {"from": "journal"}} {"content": "feature tests after the last name lookup fix came back - Import Agents and\u2026", "meta": {"from": "journal"}} {"content": "rule 38 \u2014 Never change the git branch until the user says so, by name \u2014 The work happens on the branch the user named. That was main until message 5929 and question 80 (2026-09-23), which moved the sins work to the branch sins. Do not create, switch to or merge any other branch unless the user names it in their own words.", "meta": {"from": "journal"}} {"content": "rule 35 \u2014 Write clean code - one funnel per kind of operation, never the same\u2026 \u2014 Every kind of operation has one funnel: one method that creates, one that saves, one that refuses, one that formats. A second method that does the same thing under another name splits the behaviour, and the two drift apart. Before writing a method, search for the one that already does it and extend that. scripts/checks/funnels.py finds bodies written twice.; rule 55 \u2014 Always dispatch Codex helpers on gpt-6-sol \u2014 The user's word, message 13431: switch the codex agents to GPT-6-Sol and make it their default. ~/.codex/config.toml names it as the default model too.; rule 56 \u2014 Helpers are for work that writes; subagents read, research and design \u2014 The user, message 13464: there must be a clear distinction. A subagent can be dispatched for anything read-only: research, review, design. A helper is for actual work that writes, best in its own worktree when the work is separate. Dieter designing in Claude Design should have been a subagent, not a helper.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 40 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 40 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 the files changed since the last check (`base.py`, `control\u2026 \u2014 Code Commandments \u2014 the files changed since the last check (`base.py`, `controller.py`, `controller.py`, `controller.py`, `controller.py`) breaks a rule. Fix it now, at its SOURCE, while the code is still in front of you: \u00b7 \u2022 python-blank-string-default at /Users/jessegall/projects/agent-journal/.claude/worktrees/main-edsger-lockwood/src/features/tickets/controller.py:350 \u00b7 LOAD the skill `commandments-python-absence` before fixing \u2014 load it even if you believe you already have. \u00b7 Run `commandments info ` if a rule is not one you recognise. This check reads a file at a time, so it is not the whole picture \u2014 `judge` still is.", "meta": {"from": "journal"}} {"content": "full suite for the redone action marker came back - Add the check, lint, run\u2026", "meta": {"from": "journal"}} {"content": "rule 47 \u2014 The journal sets itself up once, when the server starts, never per\u2026 \u2014 Messages 2220 and 2224. Discovering features and their handlers, seating the feature rows and the rename sweep happen once, at server boot, and again only when a feature is switched on or off, a plugin changes or an environment is added: features.load keeps a set-up generation per journal (SEATED) and redoes the work only when that generation moves. A command, a request or a hook uses what is already there; nothing in their path may rediscover handlers or rescan folders. A cost that repeats per call is a bug to fix, not a budget to raise.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 55 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 55 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "chat etiquette - a line from the journal is an instruction, not a message\u2026 \u2014 a turn that only handles a journal line needs no words: act on it, or say once in the chat what you wait on, then carry on; what the user needs to know still goes to the chat", "meta": {"from": "journal"}} {"content": "full suite for the redone controllers-by-class change came back - Full suite\u2026", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you commit \u2014 you've changed 45 judged files since th\u2026 \u2014 Code Commandments \u2014 before you commit: you've changed 45 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "rule 40 \u2014 A feature is named for what it is, never for its machinery \u2014 Messages 599, 600 and 703. A feature is a capability the user would name and would think of switching off. File tracking, a write gate, a phrase bank, a tree diff are services used inside a feature, not features of their own: they live in the feature they serve. Before adding a directory under features/, say what the user would call it; if the answer names a mechanism, it belongs inside something else. Report 16 holds the grouping this implies.", "meta": {"from": "journal"}} {"content": "the await tag does this in one step \u2014 [!await] makes the rest of the turn what you wait for; [!await on=(\"\", \"helper:\")] waits on those; it runs only when it opens the last text of your turn", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 53 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 53 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for 2608 on the newest tip came back - Full suite for 2608 on the\u2026", "meta": {"from": "journal"}} {"content": "your chat talked about the journal's workings - \"To-dos 2602 and 2608 are done\" \u2014 the user sees replies, reactions, pills and reads themselves; say what the work is instead", "meta": {"from": "journal"}} {"content": "Taken (4baf56a3). Thank you. Next for you, in this order, each a to-do (journal todo read ): 2606 one Extension type for feature extension points (clear_all on unload; tickets' leak), 2611 one stamp-keyed cache helper (engine/memo.py, forget_all for tests), 2609 layers declared in .commandments/config.json and one boot(root) per process (surfaces/updates.py -> engine/upgrades.py, agent_state into Agents), 2605 one unit of work for undo, held events and deferred jobs, and only then, if the night allows, 2604 row storage composed into controllers. Massimo Switchwell is building the Settings page (details.py and web settings files) and Barbara is narrowing report handlers in features; rebase often. Commit per to-do, report.", "meta": {"from": "main"}} {"content": "rule 42 \u2014 Every user-facing text passes the formatters before it leaves the\u2026 \u2014 Not only a brief. A title, an abstract, an outcome and every section body are read by a person, so each goes through the same formatters on its way to the viewer \u2014 chat turns, activity items, to-do rows, inspector pages, docs alike. One field formatted out of five is not a rule, it is an accident, and it is how a raw tag ended up in the activity list after the tags feature had been stripping them for weeks. When a new field carries words a person reads, it joins the list in the same place.", "meta": {"from": "journal"}} {"content": "fact 23 \u2014 Every upgrade brings system sequences and their triggers in line\u2026 \u2014 install.py runs ship_sequences after the migrations on each upgrade, so features/sequences/shipped.py is the whole source: change its wording and the next upgrade updates every journal, no migration needed. Shipped rows carry system=True and are read-only for everyone but SYSTEM (controllers/base.py _shipped).; law L5 \u2014 Every subagent dispatch names the agent - a human name, a little\u2026 \u2014 A name is how the user and the chat tell subagents apart and how they are messaged later; an id or a task line is not a name. Start the dispatch's description with the name, a colon, then the task, such as \"Dr. Einstein: profile the slow hooks\" or \"Coco Rams: draw the plan card\". A designer can borrow from famous designers, a researcher from famous scientists, mixed up for fun.", "meta": {"from": "journal"}} {"content": "law L3 \u2014 Read narrowly - grep for the line, sed a range, head the file; never\u2026 \u2014 Everything a tool returns stays in the context for good and is paid for on every turn after it. Search before you read, read the range you need, and cap output with grep, head or tail. Read a whole file only when you need all of it.", "meta": {"from": "journal"}} {"content": "rule 52 \u2014 A chat mark for something the user did sits on the user's side \u2014 Message 10960 (2026-09-25): marks for the user's own actions, such as answering a question, are right-aligned like the user's messages. A mark is put there by giving its card side=user.", "meta": {"from": "journal"}} {"content": "rule 41 \u2014 Keep moving, run the whole suite before every commit, never wait \u2014 The full suite runs in about seven seconds: .venv/bin/python -m pytest -q --timeout=300 -n auto. Run it before every commit instead of picking tests by name. Group rows that sit in the same code into one sitting: write them all, test once, commit once. And never wait, not for a subagent, a build, or an answer you can carry on without. Dispatch it and keep working. If you truly are waiting on something, say so in the work log.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 32 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 32 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for the Extension type came back - Add the unload check and run the\u2026", "meta": {"from": "journal"}} {"content": "rule 36 \u2014 Clean, DRY, idiomatic before it is committed, never after it is\u2026 \u2014 The user should never be the one who finds duplication, dead code, a clumsy name or a pattern the codebase does not use. Read the diff before every commit as a reviewer would, and fix what is not clean then, not in a follow-up after a complaint.", "meta": {"from": "journal"}} {"content": "fact 9 \u2014 Every public method on a controller becomes a journal command \u2014 The CLI is generated from the controllers: each public method of Controller, or of a typed controller, turns into journal . A helper added to the base class therefore becomes a command on every type \u2014 which is how journal handled and journal refuse came to exist, from the CRUD funnel and the refusal funnel. An internal helper on a controller is named with a leading underscore, as _shaped and _status already are, or it ships as a command nobody meant.; 12 facts standing, read them \u2014 9. Every public method on a controller becomes a journal command; 13. This live session runs the installed copy in .journal/journal.pyz; 18. cProfile inflates the slow-request profiles about tenfold; 20. A slim supervisor holds the agent and a worker reloads on every build; 23. Every upgrade brings system sequences and their triggers in line with the code; 24. An answer followed by tool calls can be missing from Claude's transcript; 25. A designer's install packs the whole tree, half-done server edits included; 26. Claude Code reads agent profiles when a session starts; 27. Running src/journal.py against the live .journal root starts a second server; 28. The designer agent type exists, so design work goes to a subagent; 29. Codex's transcript records the end of every exec session, polled or not; 30. The tunler server refuses TLS for any subdomain without a tunnel; context 50% full, decide \u2014 a fact is what a later reader would get wrong without, a rule binds every environment, or nothing \"\"; 34 rules in force, read them \u2014 7. journal disable must only ever be run because the user explicitly asked for it,; 9. Research dispatched to a subagent ends in a REPORT for the user, compiled by the; 13. A title names the thing in at most 80 characters and never explains it with a co; 14. A small journal capability is a feature under features, with at most one test; 17. Do not restart the viewer for frontend-only changes; 18. Provider-specific code belongs in providers, never features; 19. Providers report facts; features decide behavior; 22. File distinct user work requests immediately; 27. Name a declaration with the word a reader already knows; 30. The viewer has one API client, and every piece of it does one job; 31. Every finding a reviewing agent reports becomes its own to-do; 35. Write clean code - one funnel per kind of operation, never the same method twice; 36. Clean, DRY, idiomatic before it is committed, never after it is complained about; 37. Close every to-do explicitly with todo done or a Journal commit trailer; 38. Never change the git branch until the user says so, by name; 39. Use only registered exclamation response tags; 40. A feature is named for what it is, never for its machinery; 41. Keep moving, run the whole suite before every commit, never wait; 42. Every user-facing text passes the formatters before it leaves the server; 43. A request or hook over its budget is fixed before the next release; 44. Release a new version after every significant change; 45. No prose words as names in code - said, says, heard, spoke, told, shown, became; 46. Only commit and push once the whole journal is proven to boot; 47. The journal sets itself up once, when the server starts, never per command; 48. The viewer is built from its component library, and pages only compose it; 49. A dialog whose content grows keeps one fixed height, and its content scrolls; 50. Everything the user does is doable in the viewer; 51. Every finished feature is committed, pushed and released with a new version tag; 52. A chat mark for something the user did sits on the user's side; 53. Every new design gets one critique round, and the designer decides; 54. Settings and feature switches are read at boot and on change, never per call; 55. Always dispatch Codex helpers on gpt-6-sol; 56. Helpers are for work that writes; subagents read, research and design; 57. Never merge the overnight refactor into main before its pull request is approved", "meta": {"from": "journal"}} {"content": "rule 57 \u2014 Never merge the overnight refactor into main before its pull request\u2026 \u2014 Messages 15005, 15006, 15109, 15110 (2026-10-04): all refactor work goes on branch overnight-refactor and reaches the user as one pull request, which they read in the morning; nothing of it is merged into main until they say so. Hotfixes the user explicitly asks for go to main at once and are merged into the branch.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 9 judged files since th\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 9 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for the stamp-keyed memo came back - Add memo reset for tests and\u2026", "meta": {"from": "journal"}} {"content": "rule 46 \u2014 Only commit and push once the whole journal is proven to boot \u2014 Messages 2178 and 2179. A release that has not been started for real can crash every project that installs it, as 2.78.3 did for Codex and 2.84.0 did to project records. Before every commit and push: the full suite passes, including tests/test_it_boots.py, which installs a packed copy into a fresh project, launches Claude and Codex from it, writes a project record, upgrades again and checks the record survives. When a change touches launching, installing or upgrading, also start a real journal in a scratch project and close it properly afterwards, leaving no process behind.", "meta": {"from": "journal"}} {"content": "fact 25 \u2014 A designer's install packs the whole tree, half-done server edits\u2026 \u2014 2026-09-25: Eames and Saul run python3 src/journal.py --root .journal upgrade after their viewer builds; it packs every file in src, so a server handler I was halfway through writing went live and raised on every PostToolUse hook. While designers work in parallel, keep server edits whole between tool calls (write and test in the scratchpad first), and reinstall after reverting anything.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 9 judged files since th\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 9 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for the memo, rebased came back - Wait for the 2611 suite completed\u2026", "meta": {"from": "journal"}} {"content": "fact 20 \u2014 A slim supervisor holds the agent and a worker reloads on every build \u2014 Since 2.118.0 (2026-09-23). src/supervisor.py is standard library only and never reloads: journal claude hands its process over to it (os.execv), and it owns the pty and the agent process, relays the terminal, writes the printed and screen captures, listens on the typist socket, restarts the agent in the same session from a relaunch command written to its runtime folder while a restart is pending, and stops it with escalation while draining the pty (an agent cannot finish exiting on macOS while its output is unread). It starts the worker (src/worker.py, which runs runner/worker.py; engine/worker.py stays as an alias for supervisors started before 2.201) and starts it again whenever it exits: RELOAD on a new build, RELAUNCH to restart the agent, STOP to end, HEAL or a quick crash to roll back a build through journal heal. The worker holds everything else: seating the session, the start-up confirm typed through the typist, services, viewer, update check, check-in, and the one-time relaunch of sessions launched before agents/terminal.py LAUNCH. agents/terminal.py holds only journal-side helpers. The server (serve.py) still runs the engines and re-execs itself on a .py change. When the agent exits, the supervisor runs journal ended, which puts set-aside hooks back and stops the server when no session is left.", "meta": {"from": "journal"}} {"content": "rule 44 \u2014 Release a new version after every significant change \u2014 The user, message 13219 (2026-10-01): 'Don't forget to release new versions every time you do something significant.' Bump VERSION, add a CHANGELOG entry, push main and the tag. Replaces message 5929's release-on-request.; rule 50 \u2014 Everything the user does is doable in the viewer \u2014 Message 6710 (2026-09-23): the user never uses the CLI, only the UI; everything should be doable from the viewer. The journal commands are for agents; any action meant for the user (making boards, confirming, accepting, hosting, watching an agent) needs its place in the viewer.; rule 51 \u2014 Every finished feature is committed, pushed and released with a new\u2026 \u2014 Message 9207 (2026-09-24): when a new feature is ready, commit, push and publish a new tag. This is the user's standing word for releasing, so rule 44's only-when-the-user-says is met by it for finished features; fixes in between wait for the next feature or a patch the user asks for.", "meta": {"from": "journal"}} {"content": "rule 56 \u2014 Helpers are for work that writes; subagents read, research and design \u2014 The user, message 13464: there must be a clear distinction. A subagent can be dispatched for anything read-only: research, review, design. A helper is for actual work that writes, best in its own worktree when the work is separate. Dieter designing in Claude Design should have been a subagent, not a helper.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 19 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 19 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for layers and boot came back - Full suite for 2609 completed\u2026", "meta": {"from": "journal"}} {"content": "rule 38 \u2014 Never change the git branch until the user says so, by name \u2014 The work happens on the branch the user named. That was main until message 5929 and question 80 (2026-09-23), which moved the sins work to the branch sins. Do not create, switch to or merge any other branch unless the user names it in their own words.", "meta": {"from": "journal"}} {"content": "install.py names 2 files in the project \u2014 write the path so the chat can link it: install.py, src/install.py", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 18 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 18 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for layers and boot after the installer fix came back - Full suite\u2026", "meta": {"from": "journal"}} {"content": "the log tag does this in one step \u2014 [!log:N] makes the turn itself the log entry; it runs only when it opens the last text of your turn", "meta": {"from": "journal"}} {"content": "full suite for 2609 with boot in commands came back - Full suite for 2609 with\u2026", "meta": {"from": "journal"}} {"content": "rule 27 \u2014 Name a declaration with the word a reader already knows \u2014 An attribute, a variable or a field gets the ordinary programming word for what it holds, not an evocative one. was, heard and alone were poetry; aliases, notify_actions and urgent_actions are what they are. The test: could a reader who has never seen this codebase guess what it holds from the name alone? Prose belongs in the help text and the abstract, where it is read as prose. This does not license abbreviations \u2014 a plain word in full, not a short one.; rule 37 \u2014 Close every to-do explicitly with todo done or a Journal commit\u2026 \u2014 Ending work does not close its row. A to-do is closed by journal todo done --how, or by a commit whose message carries Journal: todos done at column 0, several numbers separated by commas. A row left open after its work landed misleads the next session and auto mode.", "meta": {"from": "journal"}} {"content": "rule 35 \u2014 Write clean code - one funnel per kind of operation, never the same\u2026 \u2014 Every kind of operation has one funnel: one method that creates, one that saves, one that refuses, one that formats. A second method that does the same thing under another name splits the behaviour, and the two drift apart. Before writing a method, search for the one that already does it and extend that. scripts/checks/funnels.py finds bodies written twice.; rule 47 \u2014 The journal sets itself up once, when the server starts, never per\u2026 \u2014 Messages 2220 and 2224. Discovering features and their handlers, seating the feature rows and the rename sweep happen once, at server boot, and again only when a feature is switched on or off, a plugin changes or an environment is added: features.load keeps a set-up generation per journal (SEATED) and redoes the work only when that generation moves. A command, a request or a hook uses what is already there; nothing in their path may rediscover handlers or rescan folders. A cost that repeats per call is a bug to fix, not a budget to raise.; rule 54 \u2014 Settings and feature switches are read at boot and on change, never\u2026 \u2014 The user, message 13349: the application boots, determines every feature and setting once, and re-evaluates only when something changes, such as a setting or a plugin. Never lazy-load settings.; rule 55 \u2014 Always dispatch Codex helpers on gpt-6-sol \u2014 The user's word, message 13431: switch the codex agents to GPT-6-Sol and make it their default. ~/.codex/config.toml names it as the default model too.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you commit \u2014 you've changed 14 judged files since th\u2026 \u2014 Code Commandments \u2014 before you commit: you've changed 14 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "fact 23 \u2014 Every upgrade brings system sequences and their triggers in line\u2026 \u2014 install.py runs ship_sequences after the migrations on each upgrade, so features/sequences/shipped.py is the whole source: change its wording and the next upgrade updates every journal, no migration needed. Shipped rows carry system=True and are read-only for everyone but SYSTEM (controllers/base.py _shipped).", "meta": {"from": "journal"}} {"content": "full suite for the unit of work came back - Rebase and run the full suite for\u2026", "meta": {"from": "journal"}} {"content": "stored.py names 2 files in the project \u2014 write the path so the chat can link it: src/controllers/stored.py, src/engine/stored.py; rule 42 \u2014 Every user-facing text passes the formatters before it leaves the\u2026 \u2014 Not only a brief. A title, an abstract, an outcome and every section body are read by a person, so each goes through the same formatters on its way to the viewer \u2014 chat turns, activity items, to-do rows, inspector pages, docs alike. One field formatted out of five is not a rule, it is an accident, and it is how a raw tag ended up in the activity list after the tags feature had been stripping them for weeks. When a new field carries words a person reads, it joins the list in the same place.; rule 45 \u2014 No prose words as names in code - said, says, heard, spoke, told\u2026 \u2014 Messages 1360 and 1698. The user has said more than once that code must not read like prose: a variable, attribute, property or function is named for what it holds or does (text, command, labels, lines), never with a verb from a story. 'says' on the Design type (1360) and 'said = call.said.lower()' in features/recital.py (1698) are the examples. Rule 27 states the naming rule; this one carries the words, so writing one of them whispers it. Before writing a name, ask whether a reader who has never seen the code would know what it holds.", "meta": {"from": "journal"}} {"content": "Taken: all four (2605, 2606, 2609, 2611). The review of the first three found nothing blocking; please fix its small points now: memo forget_all called from features.unload() (or deleted); Extension.each/keyed take a Record only, the Path callers (services.specs, sharing/services.py:30) build it once; the three blank lines in commands/cli.py:22-24; the lambda assigned to make in surfaces/listing.py; trigger.py's HELD.put written twice -> one helper. 2604: not yet; I will say go once Barbara's and Dieter's feature/web work has landed. Rebase first, commit, report.", "meta": {"from": "main"}} {"content": "The review of be0c6e3a found nothing blocking. Add to your current small-fixes commit: bus.held() saves and restores the previous queue instead of clobbering it (release() and WORK.after the same); undoable() yields nothing; unit() sits next to settled(); Work groups its fields by concern (or splits into bus and undo parts) with no comments - names that say it. Then report.", "meta": {"from": "main"}} {"content": "rule 41 \u2014 Keep moving, run the whole suite before every commit, never wait \u2014 The full suite runs in about seven seconds: .venv/bin/python -m pytest -q --timeout=300 -n auto. Run it before every commit instead of picking tests by name. Group rows that sit in the same code into one sitting: write them all, test once, commit once. And never wait, not for a subagent, a build, or an answer you can carry on without. Dispatch it and keep working. If you truly are waiting on something, say so in the work log.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 11 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 11 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for the review points came back - Smoke-load and run the full suite\u2026", "meta": {"from": "journal"}} {"content": "law L3 \u2014 Read narrowly - grep for the line, sed a range, head the file; never\u2026 \u2014 Everything a tool returns stays in the context for good and is paid for on every turn after it. Search before you read, read the range you need, and cap output with grep, head or tail. Read a whole file only when you need all of it.", "meta": {"from": "journal"}} {"content": "fact 9 \u2014 Every public method on a controller becomes a journal command \u2014 The CLI is generated from the controllers: each public method of Controller, or of a typed controller, turns into journal . A helper added to the base class therefore becomes a command on every type \u2014 which is how journal handled and journal refuse came to exist, from the CRUD funnel and the refusal funnel. An internal helper on a controller is named with a leading underscore, as _shaped and _status already are, or it ships as a command nobody meant.; rule 36 \u2014 Clean, DRY, idiomatic before it is committed, never after it is\u2026 \u2014 The user should never be the one who finds duplication, dead code, a clumsy name or a pattern the codebase does not use. Read the diff before every commit as a reviewer would, and fix what is not clean then, not in a follow-up after a complaint.; rule 57 \u2014 Never merge the overnight refactor into main before its pull request\u2026 \u2014 Messages 15005, 15006, 15109, 15110 (2026-10-04): all refactor work goes on branch overnight-refactor and reaches the user as one pull request, which they read in the morning; nothing of it is merged into main until they say so. Hotfixes the user explicitly asks for go to main at once and are merged into the branch.", "meta": {"from": "journal"}} {"content": "Taken (542b9d2e). Next: to-do 2667 (journal todo read 2667), the import-direction violations your declared layers still flag (features importing commands/install/serve/skills/surfaces; agents and runner importing features; runner importing commands; the remaining direct features.load() calls going through boot). Old migrations may keep importing features. Then 2604 (row storage composed), but only after Barbara reports her to-do 2630 (she is changing Resource.fork and controllers/stored.py load paths); I will tell you.", "meta": {"from": "main"}} {"content": "Barbara's 2630 has landed (51dfdb6e, 33eb28f2: Agents._primary_to_read returns the row unforked; load() still forks everywhere). Go on 2604 after 2667: row storage composed into controllers; making load fork only for writers fits in it if you want it. Script-driven as you proposed; rebase onto overnight-refactor first.", "meta": {"from": "main"}} {"content": "fact 24 \u2014 An answer followed by tool calls can be missing from Claude's\u2026 \u2014 Seen 2026-09-24 for messages 9391-9404: text blocks opening with [!reply:n] that were followed by tool calls never appeared in the session's jsonl (only thinking and tool_use rows did), so the journal never saw them and the replies were lost. When a turn goes on after answering, send the answer with journal message reply \"\" instead of the tag.; rule 48 \u2014 The viewer is built from its component library, and pages only\u2026 \u2014 Message 4258. Every visual piece the viewer shows more than once, or that a user would recognise as the same kind of thing (a dialog, a side panel or inspector, a dropdown, a list row, a switch, a button), is one component in web/src/kit, extracted aggressively, and every page composes those components instead of building its own copy. Before writing markup or styles in a page, look for the kit component that already does it and extend it with a prop; a second hand-built version is a bug. The side panel that animated in but not out, while a separate skill panel did both, is the example.", "meta": {"from": "journal"}} {"content": "rule 52 \u2014 A chat mark for something the user did sits on the user's side \u2014 Message 10960 (2026-09-25): marks for the user's own actions, such as answering a question, are right-aligned like the user's messages. A mark is put there by giving its card side=user.; law L5 \u2014 Every subagent dispatch names the agent - a human name, a little\u2026 \u2014 A name is how the user and the chat tell subagents apart and how they are messaged later; an id or a task line is not a name. Start the dispatch's description with the name, a colon, then the task, such as \"Dr. Einstein: profile the slow hooks\" or \"Coco Rams: draw the plan card\". A designer can borrow from famous designers, a researcher from famous scientists, mixed up for fun.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 30 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 30 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for the import-direction fixes came back - Re-scan violations and\u2026", "meta": {"from": "journal"}} {"content": "rule 46 \u2014 Only commit and push once the whole journal is proven to boot \u2014 Messages 2178 and 2179. A release that has not been started for real can crash every project that installs it, as 2.78.3 did for Codex and 2.84.0 did to project records. Before every commit and push: the full suite passes, including tests/test_it_boots.py, which installs a packed copy into a fresh project, launches Claude and Codex from it, writes a project record, upgrades again and checks the record survives. When a change touches launching, installing or upgrading, also start a real journal in a scratch project and close it properly afterwards, leaving no process behind.", "meta": {"from": "journal"}} {"content": "fact 25 \u2014 A designer's install packs the whole tree, half-done server edits\u2026 \u2014 2026-09-25: Eames and Saul run python3 src/journal.py --root .journal upgrade after their viewer builds; it packs every file in src, so a server handler I was halfway through writing went live and raised on every PostToolUse hook. While designers work in parallel, keep server edits whole between tool calls (write and test in the scratchpad first), and reinstall after reverting anything.", "meta": {"from": "journal"}} {"content": "rule 50 \u2014 Everything the user does is doable in the viewer \u2014 Message 6710 (2026-09-23): the user never uses the CLI, only the UI; everything should be doable from the viewer. The journal commands are for agents; any action meant for the user (making boards, confirming, accepting, hosting, watching an agent) needs its place in the viewer.", "meta": {"from": "journal"}} {"content": "rule 56 \u2014 Helpers are for work that writes; subagents read, research and design \u2014 The user, message 13464: there must be a clear distinction. A subagent can be dispatched for anything read-only: research, review, design. A helper is for actual work that writes, best in its own worktree when the work is separate. Dieter designing in Claude Design should have been a subagent, not a helper.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 142 judged files since\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 142 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for the row store came back - Lint, funnels and full suite for 2604\u2026", "meta": {"from": "journal"}} {"content": "fact 20 \u2014 A slim supervisor holds the agent and a worker reloads on every build \u2014 Since 2.118.0 (2026-09-23). src/supervisor.py is standard library only and never reloads: journal claude hands its process over to it (os.execv), and it owns the pty and the agent process, relays the terminal, writes the printed and screen captures, listens on the typist socket, restarts the agent in the same session from a relaunch command written to its runtime folder while a restart is pending, and stops it with escalation while draining the pty (an agent cannot finish exiting on macOS while its output is unread). It starts the worker (src/worker.py, which runs runner/worker.py; engine/worker.py stays as an alias for supervisors started before 2.201) and starts it again whenever it exits: RELOAD on a new build, RELAUNCH to restart the agent, STOP to end, HEAL or a quick crash to roll back a build through journal heal. The worker holds everything else: seating the session, the start-up confirm typed through the typist, services, viewer, update check, check-in, and the one-time relaunch of sessions launched before agents/terminal.py LAUNCH. agents/terminal.py holds only journal-side helpers. The server (serve.py) still runs the engines and re-execs itself on a .py change. When the agent exits, the supervisor runs journal ended, which puts set-aside hooks back and stops the server when no session is left.", "meta": {"from": "journal"}} {"content": "Taken (2a8c2abd); actions() unchanged at 1,360. Next: (1) load forks only for callers that will change the row (a read path that hands out the stored row, a write path that forks); measure GET dashboard/summary before and after on a scratch journal with curl; (2) Edsger B3.8: move each feature's routes out of commands/http.py into features//routes.py through a routes hook in features/wiring.py (the resolver no longer depends on order); http.py keeps the core row and agent routes; feature tests import the route functions from their new home. Dieter (web) and Ada/Hedy (small feature leftovers) are still working; rebase often. Commit per item, report.", "meta": {"from": "main"}} {"content": "rule 43 \u2014 A request or hook over its budget is fixed before the next release \u2014 Comment 1151 on this rule. When the faults feature reports a request, a hook or a command slower than its budget, file it as a to-do at once. It does not jump ahead of the work in hand, but no version is published while one is still open: profile it, fix it, and verify the new time before the release goes out. The budget is 50ms, because everything runs locally against files.", "meta": {"from": "journal"}} {"content": "the log tag does this in one step \u2014 [!log:N] makes the turn itself the log entry; it runs only when it opens the last text of your turn", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 5 judged files since th\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 5 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for the read path came back - Full suite for the read path\u2026", "meta": {"from": "journal"}} {"content": "fact 23 \u2014 Every upgrade brings system sequences and their triggers in line\u2026 \u2014 install.py runs ship_sequences after the migrations on each upgrade, so features/sequences/shipped.py is the whole source: change its wording and the next upgrade updates every journal, no migration needed. Shipped rows carry system=True and are read-only for everyone but SYSTEM (controllers/base.py _shipped).", "meta": {"from": "journal"}} {"content": "rule 27 \u2014 Name a declaration with the word a reader already knows \u2014 An attribute, a variable or a field gets the ordinary programming word for what it holds, not an evocative one. was, heard and alone were poetry; aliases, notify_actions and urgent_actions are what they are. The test: could a reader who has never seen this codebase guess what it holds from the name alone? Prose belongs in the help text and the abstract, where it is read as prose. This does not license abbreviations \u2014 a plain word in full, not a short one.; rule 44 \u2014 Release a new version after every significant change \u2014 The user, message 13219 (2026-10-01): 'Don't forget to release new versions every time you do something significant.' Bump VERSION, add a CHANGELOG entry, push main and the tag. Replaces message 5929's release-on-request.; rule 51 \u2014 Every finished feature is committed, pushed and released with a new\u2026 \u2014 Message 9207 (2026-09-24): when a new feature is ready, commit, push and publish a new tag. This is the user's standing word for releasing, so rule 44's only-when-the-user-says is met by it for finished features; fixes in between wait for the next feature or a patch the user asks for.; rule 54 \u2014 Settings and feature switches are read at boot and on change, never\u2026 \u2014 The user, message 13349: the application boots, determines every feature and setting once, and re-evaluates only when something changes, such as a setting or a plugin. Never lazy-load settings.", "meta": {"from": "journal"}} {"content": "rule 47 \u2014 The journal sets itself up once, when the server starts, never per\u2026 \u2014 Messages 2220 and 2224. Discovering features and their handlers, seating the feature rows and the rename sweep happen once, at server boot, and again only when a feature is switched on or off, a plugin changes or an environment is added: features.load keeps a set-up generation per journal (SEATED) and redoes the work only when that generation moves. A command, a request or a hook uses what is already there; nothing in their path may rediscover handlers or rescan folders. A cost that repeats per call is a bug to fix, not a budget to raise.", "meta": {"from": "journal"}} {"content": "rule 40 \u2014 A feature is named for what it is, never for its machinery \u2014 Messages 599, 600 and 703. A feature is a capability the user would name and would think of switching off. File tracking, a write gate, a phrase bank, a tree diff are services used inside a feature, not features of their own: they live in the feature they serve. Before adding a directory under features/, say what the user would call it; if the answer names a mechanism, it belongs inside something else. Report 16 holds the grouping this implies.", "meta": {"from": "journal"}} {"content": "law L3 \u2014 Read narrowly - grep for the line, sed a range, head the file; never\u2026 \u2014 Everything a tool returns stays in the context for good and is paid for on every turn after it. Search before you read, read the range you need, and cap output with grep, head or tail. Read a whole file only when you need all of it.", "meta": {"from": "journal"}} {"content": "rule 41 \u2014 Keep moving, run the whole suite before every commit, never wait \u2014 The full suite runs in about seven seconds: .venv/bin/python -m pytest -q --timeout=300 -n auto. Run it before every commit instead of picking tests by name. Group rows that sit in the same code into one sitting: write them all, test once, commit once. And never wait, not for a subagent, a build, or an answer you can carry on without. Dispatch it and keep working. If you truly are waiting on something, say so in the work log.", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 26 judged files since t\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 26 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "full suite for the feature routes came back - Load features in the shadow\u2026", "meta": {"from": "journal"}} {"content": "rule 36 \u2014 Clean, DRY, idiomatic before it is committed, never after it is\u2026 \u2014 The user should never be the one who finds duplication, dead code, a clumsy name or a pattern the codebase does not use. Read the diff before every commit as a reviewer would, and fix what is not clean then, not in a follow-up after a complaint.; rule 57 \u2014 Never merge the overnight refactor into main before its pull request\u2026 \u2014 Messages 15005, 15006, 15109, 15110 (2026-10-04): all refactor work goes on branch overnight-refactor and reaches the user as one pull request, which they read in the morning; nothing of it is merged into main until they say so. Hotfixes the user explicitly asks for go to main at once and are merged into the branch.", "meta": {"from": "journal"}} {"content": "fact 9 \u2014 Every public method on a controller becomes a journal command \u2014 The CLI is generated from the controllers: each public method of Controller, or of a typed controller, turns into journal . A helper added to the base class therefore becomes a command on every type \u2014 which is how journal handled and journal refuse came to exist, from the CRUD funnel and the refusal funnel. An internal helper on a controller is named with a leading underscore, as _shaped and _status already are, or it ships as a command nobody meant.; rule 35 \u2014 Write clean code - one funnel per kind of operation, never the same\u2026 \u2014 Every kind of operation has one funnel: one method that creates, one that saves, one that refuses, one that formats. A second method that does the same thing under another name splits the behaviour, and the two drift apart. Before writing a method, search for the one that already does it and extend that. scripts/checks/funnels.py finds bodies written twice.; rule 37 \u2014 Close every to-do explicitly with todo done or a Journal commit\u2026 \u2014 Ending work does not close its row. A to-do is closed by journal todo done --how, or by a commit whose message carries Journal: todos done at column 0, several numbers separated by commas. A row left open after its work landed misleads the next session and auto mode.; rule 55 \u2014 Always dispatch Codex helpers on gpt-6-sol \u2014 The user's word, message 13431: switch the codex agents to GPT-6-Sol and make it their default. ~/.codex/config.toml names it as the default model too.", "meta": {"from": "journal"}} {"content": "Taken (both). Thank you; that is the core list done. Stand by for the final review's follow-ups.", "meta": {"from": "main"}} {"content": "Next for you (rebase first): to-do 2664 (journal todo read 2664), one elected owner for engine supervision so Children can move from the server into the worker (a flock like Engines.run), only if it stays safe; otherwise report why it should stay in the server. Report by 08:00.", "meta": {"from": "main"}} {"content": "the log tag does this in one step \u2014 [!log:N] makes the turn itself the log entry; it runs only when it opens the last text of your turn", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you wrap up \u2014 you've changed 3 judged files since th\u2026 \u2014 Code Commandments \u2014 before you wrap up: you've changed 3 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "chat etiquette - a line from the journal is an instruction, not a message\u2026 \u2014 a turn that only handles a journal line needs no words: act on it, or say once in the chat what you wait on, then carry on; what the user needs to know still goes to the chat", "meta": {"from": "journal"}} {"content": "full suite for the elected engine supervisor came back - Full suite for the\u2026", "meta": {"from": "journal"}} {"content": "your command ran 32s in the foreground and was moved to the background \u2014 carry on with other work; you are told when it ends. Start a command you expect to take long in the background yourself", "meta": {"from": "journal"}} {"content": "rule 48 \u2014 The viewer is built from its component library, and pages only\u2026 \u2014 Message 4258. Every visual piece the viewer shows more than once, or that a user would recognise as the same kind of thing (a dialog, a side panel or inspector, a dropdown, a list row, a switch, a button), is one component in web/src/kit, extracted aggressively, and every page composes those components instead of building its own copy. Before writing markup or styles in a page, look for the kit component that already does it and extend it with a prop; a second hand-built version is a bug. The side panel that animated in but not out, while a separate skill panel did both, is the example.", "meta": {"from": "journal"}} {"content": "12 facts standing, read them \u2014 9. Every public method on a controller becomes a journal command; 13. This live session runs the installed copy in .journal/journal.pyz; 18. cProfile inflates the slow-request profiles about tenfold; 20. A slim supervisor holds the agent and a worker reloads on every build; 23. Every upgrade brings system sequences and their triggers in line with the code; 24. An answer followed by tool calls can be missing from Claude's transcript; 25. A designer's install packs the whole tree, half-done server edits included; 26. Claude Code reads agent profiles when a session starts; 27. Running src/journal.py against the live .journal root starts a second server; 28. The designer agent type exists, so design work goes to a subagent; 29. Codex's transcript records the end of every exec session, polled or not; 30. The tunler server refuses TLS for any subdomain without a tunnel; 34 rules in force, read them \u2014 7. journal disable must only ever be run because the user explicitly asked for it,; 9. Research dispatched to a subagent ends in a REPORT for the user, compiled by the; 13. A title names the thing in at most 80 characters and never explains it with a co; 14. A small journal capability is a feature under features, with at most one test; 17. Do not restart the viewer for frontend-only changes; 18. Provider-specific code belongs in providers, never features; 19. Providers report facts; features decide behavior; 22. File distinct user work requests immediately; 27. Name a declaration with the word a reader already knows; 30. The viewer has one API client, and every piece of it does one job; 31. Every finding a reviewing agent reports becomes its own to-do; 35. Write clean code - one funnel per kind of operation, never the same method twice; 36. Clean, DRY, idiomatic before it is committed, never after it is complained about; 37. Close every to-do explicitly with todo done or a Journal commit trailer; 38. Never change the git branch until the user says so, by name; 39. Use only registered exclamation response tags; 40. A feature is named for what it is, never for its machinery; 41. Keep moving, run the whole suite before every commit, never wait; 42. Every user-facing text passes the formatters before it leaves the server; 43. A request or hook over its budget is fixed before the next release; 44. Release a new version after every significant change; 45. No prose words as names in code - said, says, heard, spoke, told, shown, became; 46. Only commit and push once the whole journal is proven to boot; 47. The journal sets itself up once, when the server starts, never per command; 48. The viewer is built from its component library, and pages only compose it; 49. A dialog whose content grows keeps one fixed height, and its content scrolls; 50. Everything the user does is doable in the viewer; 51. Every finished feature is committed, pushed and released with a new version tag; 52. A chat mark for something the user did sits on the user's side; 53. Every new design gets one critique round, and the designer decides; 54. Settings and feature switches are read at boot and on change, never per call; 55. Always dispatch Codex helpers on gpt-6-sol; 56. Helpers are for work that writes; subagents read, research and design; 57. Never merge the overnight refactor into main before its pull request is approved", "meta": {"from": "journal"}} {"content": "boot test reruns after the worker SIGTERM handling came back - Handle SIGTERM\u2026", "meta": {"from": "journal"}} {"content": "the await tag does this in one step \u2014 [!await] makes the rest of the turn what you wait for; [!await on=(\"\", \"helper:\")] waits on those; it runs only when it opens the last text of your turn", "meta": {"from": "journal"}} {"content": "four launch boot test reruns came back - Rerun the launch boot test four times\u2026", "meta": {"from": "journal"}} {"content": "your command ran 34s in the foreground and was moved to the background \u2014 carry on with other work; you are told when it ends. Start a command you expect to take long in the background yourself", "meta": {"from": "journal"}} {"content": "rule 46 \u2014 Only commit and push once the whole journal is proven to boot \u2014 Messages 2178 and 2179. A release that has not been started for real can crash every project that installs it, as 2.78.3 did for Codex and 2.84.0 did to project records. Before every commit and push: the full suite passes, including tests/test_it_boots.py, which installs a packed copy into a fresh project, launches Claude and Codex from it, writes a project record, upgrades again and checks the record survives. When a change touches launching, installing or upgrading, also start a real journal in a scratch project and close it properly afterwards, leaving no process behind.", "meta": {"from": "journal"}} {"content": "full suite for the elected engine supervisor, rebased came back - Full suite\u2026", "meta": {"from": "journal"}} {"content": "Code Commandments \u2014 before you commit \u2014 you've changed 3 judged files since the\u2026 \u2014 Code Commandments \u2014 before you commit: you've changed 3 judged files since the last commit. Consider running `commandments judge --changes` to confirm they conform, and fix any sin at its SOURCE (don't launder a finding with a default/cast/null-check). This is a one-time nudge for this batch \u2014 if you've already judged, or these changes aren't worth a scan, just say so and carry on.", "meta": {"from": "journal"}} {"content": "commandments-python-dependency-direction, commandments-python-value-objects\u2026 \u2014 load one again when you next need it; only the every-start skills are held for; rule 56 \u2014 Helpers are for work that writes; subagents read, research and design \u2014 The user, message 13464: there must be a clear distinction. A subagent can be dispatched for anything read-only: research, review, design. A helper is for actual work that writes, best in its own worktree when the work is separate. Dieter designing in Claude Design should have been a subagent, not a helper.", "meta": {"from": "journal"}}