Commit Graph
1179 Commits
Author SHA1 Message Date
bchanot f8665b292b chore(model-router): routing.json — confirm 7 agent rows, reflect/plan/apply phases, gitflow + ship-feature skills 2026-10-16 18:36:36 +02:00
bchanot 28993c15e0 Merge chore/impeccable-build-path into develop 2026-10-16 18:20:05 +02:00
bchanot f08efa3b9a chore(design): impeccable is the Build path, gate blocks on it instead of frontend-design
GATE-BLOCK swaps frontend-design for impeccable; Build line, gate prose, advisor rows follow. Detector rule count 45 to 61. BDR-119.
2026-10-16 18:19:29 +02:00
bchanot 218fc86494 Merge feature/short-commit-messages into develop 2026-10-16 11:49:28 +02:00
bchanot e91dcb3f12 feat(commits): follow project convention, else short messages
New lib/commit-message.md looks for a documented convention first. Without one, messages stay under 50 words at medium technical level.
2026-10-16 11:49:26 +02:00
bchanot eb1cfe55a9 Merge branch 'develop' of ssh://git.bchanot.fr:49222/bchanot/claude into develop 2026-10-15 00:12:57 +02:00
bchanot 84e3511b30 Merge bugfix/router-probe-before-route into develop 2026-10-15 00:12:23 +02:00
bchanot 304270fe5b chore(memory): BLK-030, LRN-218, TODO + journal, router-probe-before-route contract + plan 2026-10-15 00:12:18 +02:00
bchanot da0b4a65ba docs: model-router move-target probe, probe key, stand-in effort — bugfix router-probe-before-route 2026-10-15 00:12:10 +02:00
bchanot 2cccc43d37 fix(model-router): probe a move target before routing main to it
The breaker learned availability only from a failed request: with
Fable out of usage credits, every new session and every hold lapse
routed main to Fable and lost the turn to a 429
(model_requires_usage_credits, typed rate_limit).

decideFor now probes the model a decision moves main to, once per
session, with a 1-call $.model.complete raced against 8 s: a timeout
or rejection marks it down (reason "probe") and re-decides once. The
current model and held ids are never probed; `probe: false` in
~/.claude/model-router.json turns it off. When a model ranked below
its tier head stands in where the head would have run, the phase's
own effort goes one notch up (reflect xhigh, plan max on Opus).
2026-10-15 00:07:10 +02:00
bchanot dc6507dc86 Merge chore/self-explanatory-code into develop 2026-10-14 12:51:40 +02:00
bchanot 3065f08328 chore(rules): self-explanatory code, literal-scope guideline 2026-10-14 12:51:39 +02:00
bchanot fce24644b4 Merge feature/router-downgrade-cap into develop 2026-10-13 21:59:39 +02:00
bchanot aaeb125490 docs: model-router general-purpose pin, prompt estimate, engine cache TTL — feat router-cap-residuals 2026-10-13 21:37:52 +02:00
bchanot 2b91ce26bd chore(memory): BDR-116 amendment 2, LRN-217, TODO + journal, router-cap-residuals contract + plan 2026-10-13 21:37:30 +02:00
bchanot 1be57a382c feat(model-router): close the cap residuals: session pin for inheriting spawns, prompt-size estimate, engine cache TTL
A built-in general-purpose agent with no explicit model, spawned by main
while main sits below the session model, is sent the session model
instead of inheriting the demoted one (narrow: built-in source, no fork,
no workflow, main's own child; other types carry their own frontmatter
model the hook cannot see). The cold-read cap now counts the prompt the
engine's usage may not include yet: code points / 3 of the submitted
text, its context entries and main's typed skill body, summed, added at
the turn's first main step unless usage already grew by half of it;
a document attachment makes the size unknown (blocks). Texts and the
step read the same adjusted size; the served record stays raw. The
cache TTL follows the engine's cache_ttl ('5m' | '1h') from main's
PostModelSwitch events, survives /clear, and /route show prints it.
2026-10-13 19:56:17 +02:00
bchanot ab963e97dc chore(memory): BDR-116 amendment, LRN-216, EVAL-047, TODO + journal, router-downgrade-cap contract + plan 2026-10-13 09:42:47 +02:00
bchanot e8186409df docs: model-router cold-read caps, switch on by default, classifier routes down — feat router-downgrade-cap 2026-10-13 09:42:46 +02:00
bchanot 2ce203a2da feat(model-router): cold-read cap on main-loop downgrades, switch on by default, classifier may route down
The main loop may now move down to a cheaper model by default
(mainModelSwitch true, fail-closed picker). Both caps measure the COLD
READ a switch costs: the context minus what the target model's prompt
cache already holds, from a per-model served record written on every
main step (1 h TTL, cleared by /clear and /route reload; a smaller
context than the record counts as cold). downgradeMaxTokens defaults to
120000, upgradeMaxTokens keeps 200000; no return bypass is needed since
a warm return is cheap by construction.

The prompt classifier writes the full phase route: coding and
bookkeeping guesses move main down under the switch and the cap; a guess
never ranks above the session model; a mid-turn ultrathink floor keeps
its model; a user /effort typed between turns drops the guess for that
turn only (engine baseline is now per turn). A model-loaded helper skill
inside a best-tier run moves effort only. Main routed onto a model whose
window the context outgrew returns to the session model. The skip line
is said once per cap crossing; the toast names a cap hold; the status
line shows the switch only when off.
2026-10-13 01:13:06 +02:00
bchanot f3b47287f9 Merge chore/rust-default into develop 2026-10-12 23:31:19 +02:00
bchanot 1287810d15 chore(memory): BDR-118 Rust default language 2026-10-12 23:31:18 +02:00
bchanot ee767239f2 chore(rules): Rust as default language for new projects and whole-project rewrites
rules/default-language.md loads always-on (CLAUDE.global.md at 317/320,
BDR-085 exception path). Interviewer proposes Rust when the language is
not stated; init-project captures the language choice as a founding BDR.
2026-10-12 23:03:02 +02:00
bchanot 95f87cf4e3 Merge chore/code-clean into develop 2026-10-12 22:56:20 +02:00
bchanot 8693fff1d8 chore(memory): code-clean run, LRN-213/214/215, EVAL-046, BDR-117, journal, TODO, plan + audits 2026-10-12 22:51:53 +02:00
bchanot 84446612ae refactor(toggle-external,profile): one tool roster, split arms, shared gstack helpers
code-clean B5 (enable_tool split into enable_gstack/
enable_external_single/enable_pack), B7 (enable_skill into
enable_gstack_skill/enable_linked_skill/enable_plugin_skill), B9
(cmd_show into _show_entries/_show_plain/_show_grouped, `_show_plain`
ends with an explicit return 0), B18 (locals), B19 (comment).

/refactor C4: tool_kind/tool_source replace the 3 hand-copied name
lists with the MANAGED_TOOLS roster (exact-match loop), and disable_* is
split to mirror enable_*.

/refactor C5: gstack_skills/gstack_parked_names/gstack_park/
gstack_unpark move into lib/gstack-removed.sh, the only sourced file the
fixture suites copy, and profile.sh and toggle-external.sh both use
them.

496-run golden capture identical; make test 47 green, shellcheck 0.
2026-10-12 22:51:41 +02:00
bchanot 4c966ebdfc fix(toggle-external): --help prints the whole header, all 14 tools
usage() printed lines 3-26 with a hardcoded range, but the tool list
runs to line 32, so 8 of the 14 tools were missing. It now reads
`sed -n '3,/^# ====/p' | sed '$d'` and stops at the closing banner, with
no line numbers. Logged in .claude/audits/BUGS-FOUND.md.
2026-10-12 22:51:37 +02:00
bchanot 381ed9fa5d chore(code-clean): lint directives, guarded cd, hook regex split, helper extractions
A1 drop unused `import gzip`; B1/B2 file-level shellcheck directives
placed before `set -u`; B3 `cd "$REPO" || exit 1`; B4 hook regex split
into `pattern+=` lines (sha256 of $pattern unchanged); B11
`_build_parser`/`_dispatch` pulled out of google_seo `_cli`; B20
detect-plugins header names update-all.sh; C2 prune-memory deterministic
suite wired into SUITES (literal path); C3 `_refuse_outside_allowlist`
moved up in deploy-commit.

No behavior change except B3 (a failed cd now exits 1) and C2 (one more
suite). B5/B7/B9/B18/B19 land in the refactor commit.
2026-10-12 22:51:28 +02:00
bchanot d08929d0e8 Merge bugfix/model-router-residuals into develop
# Conflicts:
#	.claude/tasks/TODO.md
2026-10-11 23:42:47 +02:00
bchanot 0efcdaa1f6 Merge bugfix/push-guard-argv into develop 2026-10-11 23:42:21 +02:00
bchanot a72f360b31 chore(memory): journal — push-guard argv + model-router residual sweep 2026-10-11 23:39:18 +02:00
bchanot 85ef5e11ca chore(model-router): routing.json — analyzer confirmed on judge (live Keep), stale security-auditor changed entry purged 2026-10-11 23:39:02 +02:00
bchanot a3fbb9ae3f fix(model-router): close the accepted residuals of waves 1-4
Kill switch fails closed: a present-but-unreadable override at first load
or a non-boolean enabled turns the router off (logged); /route reload says
when the previous config was kept. Prompt rules with nested quantifiers
are dropped; a config path that is not a regular file reads as failed;
caught errors log kind + code only (errWord); rejected keys are masked in
logs (keyWord), alias keys match KEY_NAME, prototype names are never keys;
offers capped at 256; a models alias without a fallback list is ranked at
the end of the default chain; canonical matches a table id at a '-'
boundary only; logCall/noteMain cannot unroute a step; route-tool failure
text no longer claims nothing was routed.

Repo side: mods suite probes the 'claude plugin test' usage line, doctor
bounds plugin list to 60 s, update-all skips @skills-dir mods.
Kit 431/431, full make test green.
2026-10-11 23:39:01 +02:00
bchanot fe7d5cdb28 fix(push-guard): argv check — a push must be git's own subcommand
The text regex becomes a prefilter. The command is split into simple
commands by an awk tokenizer (quotes, escapes, ; && || | ( ) backticks,
heredoc bodies dropped) and a push is only a simple command whose argv[0]
is git (or */git) with push or send-pack as its first non-option word.
grep "git push", git log -S, git config --get push.default, git stash
push, heredocs and commit messages quoting a push are no longer denied.
bash|sh|zsh -c scripts are re-checked (depth 3); wrappers (sudo, env,
eval, xargs, timeout, git subtree...), assignments whose value reads as
a push, a missing awk or > 2000 simple commands keep the text verdict.
Prefilter also matches "git" push (quote characters stripped).
Suite 98 -> 124 checks.
2026-10-11 23:24:39 +02:00
bchanot 37ec8e332a Merge feature/model-router-w4 into develop 2026-10-11 21:39:20 +02:00
bchanot 5389ee53e7 chore(memory): BDR-116 amendment (classifier) + EVAL-045 + journal/TODO/contract w4b — feat model-router wave 4-B 2026-10-11 21:07:04 +02:00
bchanot e54a2b5d61 feat(model-router): wave 4-B — prompt classifier
Plain composer prompts with no route in force get a turn-scoped effort
from $.model.classify (design/analysis/coding/bookkeeping/chat → phase
effort, effort only, never the model). Bounded by a 3 s clock race, a
user interrupt neither counts nor writes, three failures in a row pause
it for cooldownMinutes (one toast, status marker, cleared by /route
reload). No dialog, no routing.json write on a classified route. New
Config key `classify` (default on, pickSwitch), `/route classify on|off`
session toggle, `/route show` line and argumentHint. An explicit user
/effort beats the guess (engineBase check).

Docs: README classifier paragraph, USAGE row + sentence, CHANGELOG,
effort-shift.md. Kit suite 428 tests.
2026-10-11 21:05:29 +02:00
bchanot 4506e0711f chore(memory): BDR-116 amendment (transition logs) + journal/TODO/contract w4a — feat model-router wave 4-A 2026-10-11 18:58:04 +02:00
bchanot 86845fcac8 chore(model-router): routing.json decisions — plan-challenger kept on judge, judge phase kept (live dialogs, user Keep) 2026-10-11 18:58:03 +02:00
bchanot 68d7ea65a2 feat(model-router): wave 4-A — transition-only logs, residual hardening
verbose now logs one line per real change, built from the decision
itself: main 'fable/high → fable/xhigh (plan: skill /ship-feature)' once
per turn per transition, a hand-off line per agent dispatch deduped per
(type, target) 'main fable/high → agent verifier sonnet/xhigh (row verify,
row)', a later agent effort move, and an answered-by anomaly for agents.
The four stream lines (step, answered by, spawn, route JSON, skill path)
are gone; the host prefixes the plugin name, so the new lines carry none.
Lines are built inside safely before next (a log failure never unroutes a
step); names are cleaned (≤ 40 code points). Residuals: failure texts go to
the log with a generic answer (forget, ask), file fragments cleaned at
render, floorFile name allowlisted, __proto__ refused in restoreOf, the
changed scan limited to rows, dialog Change offers only shipped phases,
a post-answer write failure keeps the key asked. Kit suite 284 → 358.

Contract .claude/tasks/contracts/2026-10-11-model-router-w4a-logs-residuals-1637.md,
plan r3: 3 lenses (BLOCKER: agent model moves are decided at spawn) + 1
confirmation, feater + 4 rounds, GATE 0 MET, verifier at the cap on
coverage (user-accepted), security PASS + 2 MEDIUM fixed.
2026-10-11 18:56:14 +02:00
bchanot f1cea54f41 Merge feature/model-router-confirm into develop 2026-10-11 16:20:01 +02:00
bchanot 119c2e2fc8 chore(memory): BDR-116 amendment (forget) + journal/TODO/contract w3c — feat model-router wave 3-C 2026-10-11 16:15:29 +02:00
bchanot e31db7b3f3 chore(model-router): routing.json decision — escalate phase kept (live first-use dialog, user Keep) 2026-10-11 16:15:29 +02:00
bchanot cd8d72f01f feat(model-router): wave 3-C — /route forget <name|all|projects>
Clears first-use decisions from the tracked routing.json through the
existing writer: a name is forgotten in every table (confirmed, changed
with the row restored to its recorded shipped phase under guards, project
exceptions pruned) and asked again; all and projects ask a confirmation in
the engine dialog (Cancel first) and hold the single-dialog slot; nothing
is written when there is nothing to forget; the answer names what was
restored and the frontmatter floor to realign when one was aligned; the
route tool has no forget path. Docs name the new writer. Kit suite 232 → 284.

Contract .claude/tasks/contracts/2026-10-11-model-router-w3c-forget-1457.md,
plan r3: 3 lenses + 1 confirmation, feater + 3 rounds (one real defect),
GATE 0 MET, verifier at the cap on coverage (user-accepted), security PASS.
2026-10-11 16:14:59 +02:00
bchanot 563a154446 chore(memory): BDR-116 amendment + EVAL-044 + journal/TODO/contract w3b — feat model-router wave 3-B 2026-10-11 14:25:53 +02:00
bchanot 70416222b1 docs: first-use dialog with context (README, USAGE, ARCHITECTURE, CHANGELOG), security-auditor on opus — feat model-router wave 3-B 2026-10-11 14:25:53 +02:00
bchanot 938c3d1a67 chore(agents): security-auditor floor aligned to its judge row (model: opus); model-routing lock updated 2026-10-11 14:25:45 +02:00
bchanot 32b71d25b5 chore(model-router): routing.json decision — security-auditor on judge (opus/xhigh), user's answer to the new first-use dialog; changed.from restored 2026-10-11 14:11:08 +02:00
bchanot 604a6c4411 feat(model-router): wave 3-B — first-use dialog with context, model then effort on change
The dialog now explains what it asks about: the skill's description (from
its SKILL.md frontmatter, five YAML forms, first sentence, cleaned), the
agent's description (from agent.offer), the phase's new 'about' line in
routing.json, and the real model id and effort the next step runs on.
Options Later / Keep / Change; Change asks the model (fable, opus, sonnet,
haiku with their tier role), then the effort among those the phases of
that model offer, then the scope; the pair maps to an existing phase (rows
stay phase names; the row's current phase wins a tie; same phase = Keep).
A main-loop phase (T3) is Later / Keep only. A main-row change toasts the
real decision and the /route switch hint when the pick is a downgrade.
Descriptions pass a hardened read (name allowlist, stat kind, size cap)
and clean(). Kit suite 190 → 232; W3-A dialog tests migrated.

Contract .claude/tasks/contracts/2026-10-11-model-router-w3b-dialog-1240.md,
plan r3: 3 lenses (2 BLOCKERs: inline rows, phase edits) + 1 confirmation,
feater + 2 rounds, GATE 0 MET, verifier CONFORME (3rd pass), security PASS.
2026-10-11 14:10:24 +02:00
bchanot f2f404a002 chore(memory): BDR-116 + LRN-212 + EVAL-043 + journal/TODO/contract w3a — feat model-router wave 3-A 2026-10-11 11:57:56 +02:00
bchanot 5bab61e13a chore(model-router): routing.json decision — doc-syncer kept on write (live first-use dialog, user Keep) 2026-10-11 11:57:56 +02:00