Commit Graph
1323 Commits
Author SHA1 Message Date
Huanyi Xie ad7385b9a5 fix(orchestration): use supported Codex worker command 2026-10-06 10:33:25 +03:00
Yann Roberto d69eaecf25 fix(state-store): keep cleanup diagnostics off caller errors 2026-10-06 00:36:11 +08:00
sybertoooth7-lgtm 8c560ac0cd fix(ci): close catalog gate holes on README project tree
The catalog gate validated only the first README project-tree block and
had no skills/commands counterpart to the agents pattern, so:

- README.md project tree (skills) and (commands) were never checked.
- A stale count in a second tree block was invisible because the agents
  pattern used String.match with a non-global regex, which reads only the
  first occurrence.

Both are now fail-closed. parseReadmeExpectations walks every tree block
for all three categories via matchAll, reports the offending line number
in the source label, and throws when an entry is missing entirely rather
than skipping it. syncEnglishReadme uses replaceAllOrThrow so --write
repairs every block instead of only the first; replaceOrThrow could not
be reused because it replaces only the first match and its regex.test()
probe is unsafe once the pattern carries the g flag.

README.md:847 documented 67 agents against an actual 68 and was never
checked, so it was corrected.

Fixtures in tests/ci/catalog.test.js and tests/ci/validators.test.js now
emit the skills and commands tree lines the gate requires, and both gain
regression coverage that fails against the previous parser (5 new tests).
2026-10-05 19:20:13 +03:00
L4XB f838085274 fix(gateguard): reach SQL clients behind shell flag clusters, su and taskset
main resolves SQL clients behind env -S and its launcher grammars, but a
destructive statement still passed under four forms:

- a shell -c inside a short-option cluster (bash -lc, sh -ec), since only
  an exact -c was looked up;
- su, whose -c / --command / -lc value runs through the target user's
  shell, before or after --;
- taskset, whose mask or CPU list precedes the command, like timeout's
  duration;
- a command nested past the shell -c recursion limit, which was allowed
  instead of denied.

The first three are checked like sh -c; past the limit the check now
fails closed.
2026-10-05 15:56:53 +02:00
montjeffrey 6dbf6e6883 fix(gateguard): report fresh-process failures and fail the gate on them
The previous fix stopped a throwing hook from aborting the cold replay by
recording it as an allow, matching the warm path. But coldSummary only counted
disagreements, so a failure on a step the worker also allowed was no
disagreement and left no trace: the report showed no failure, a run with no
measured step reported a cold latency of 0 ms because percentile of an empty
list is 0, and the exit status passed. Trading a crash for a clean-looking
success hid exactly the failure the crash had exposed.

coldSummary now counts errors and samples, and reports no latency when nothing
was measured rather than a 0 that reads as a fast hook. The report prints "no
valid sample" in place of a latency and adds a line for steps that failed in a
fresh process, beside the existing disagreement line. The exit check moves into
gateFails, which fails on cold errors as it already did on warm errors and cold
disagreements, and is exported with coldSummary so both can be tested.

The render test builds its report from summarize([]), the real summary shape,
so it keeps working as metrics are added.
2026-10-04 22:34:30 -04:00
montjeffrey 3a4a574609 fix(gateguard): a throwing hook no longer ends the fresh-process replay
measureColdLatency ran each step in a fresh process through execFileSync,
inside try/finally with no catch. A hook that threw made the child exit
non-zero, execFileSync threw, and the error left measureColdLatency and then
evaluate, so neither the report nor the requested SARIF was written for a run
that had hit a single bad step.

The probe now catches a throwing require or run itself and records it the way
workerMain does: the raw payload stands in for the result, so the step reads as
an allow, and the error is kept. Matching the warm path matters because
coldSummary counts every cold decision that differs from the warm one as a
disagreement; any other choice would have reported disagreements that do not
exist. A child that dies without writing parseable output is caught outside the
probe and recorded the same way with no latency, and coldSummary leaves
unmeasured steps out of the percentiles while still counting them.

The new tests run a hook that always throws through both passes. They fail
against the previous code, 2 of 17, and pass against this one.
2026-10-04 22:22:50 -04:00
montjeffrey 56a516c2c5 Merge upstream/main into feat/gateguard-fewer-round-trips
Brings in the three commits main gained since the branch's merge base: the 2.2.3
release, retiring the Everything Claude Code name for ECC, and the ja-JP sponsor
update. 227 files arrive from main; two conflicted.

CHANGELOG.md: the branch had added an Unreleased section and main a released
2.2.3 section, both at the top. Both are kept verbatim with Unreleased above
2.2.3, since none of the branch's entries shipped in that release.

manifests/context-packs/skill-triggers@1.json: both sides had recomputed
registryDigest from the same earlier value to different results, and neither is
the digest of the merged tree. It is recomputed with loadContextRegistry over the
merged tree and only that field changes; triggers and triggersDigest are
untouched on both sides. validate-context-profiles passes.

All twelve CI validators pass. The full suite reports 44 failures in seven files;
43 of them fail by name and count identically on an unmodified upstream/main, and
the remaining one, plugin-hook-bootstrap, fails only inside run-all and passes in
isolation on both trees. The merge introduces no failing test that main does not
already have.
2026-10-04 20:47:30 -04:00
montjeffrey 8a7d61299c feat(gateguard): ask what outside the repository fixes a format or semantics
Across 240 evaluation trials every question the gate raised was about code
structure - importers, callers, public API, data schemas - while every deciding
fact in the hidden-intent corpus was an external contract: a consumer's CSV
dialect, a reporting timezone, retry safety against a non-idempotent endpoint,
collation rules, a quota policy. Obtaining the deciding fact predicted the
outcome almost perfectly in both arms, and the gate did not reliably cause it.

Add an external-contract question for code targets whose change profile touches
data, beside data-schema rather than replacing it: data-schema asks what the
change touches, external-contract asks who decides what it has to be. It reuses
the existing touchesData signal, so it stays as selective as the question it
sits with and adds no new detection, and it therefore does not reach a change
with no data involvement - pure retry semantics, for instance - which is a known
limit to be measured rather than guessed at.

Code edits that touch data now ask four questions instead of three, so the
condensed hint for an unknown profile changes too. That string was pinned by a
test as legacy-stable; the pin is updated deliberately, because a hint that
names fewer questions than the gate asks is worse than a hint that changed.

Suites: target-class 31, fact-force 1019, scenario-gate 25, eval 15, and the two
effectiveness suites 29 and 26, all passing. The 23 relevantQuestions
annotations that listed data-schema now list external-contract as well.
2026-10-04 18:13:02 -04:00
Yann Roberto 8be6a6f292 chore: sync state store branch with main 2026-10-05 00:15:25 +08:00
Yann Roberto d38b079115 chore: sync session alias branch with main 2026-10-05 00:15:20 +08:00
Yann Roberto 159440d259 chore: sync context routing branch with main 2026-10-05 00:13:45 +08:00
Md. Kishor MorolandKishor Morol bfaa53935a Merge upstream/main into feat/bengali-language-support
Keep both the new Polski locale and Bengali in every language switcher,
locale list and alias map; cross-link docs/pl and docs/bn. Resync catalog
counts (69 agents, 294 skills) and the skill-trigger registry digest.

Co-Authored-By: Kishor Morol <kishormorol@gmail.com>
2026-10-03 19:54:18 -04:00
basil-k-aji-dev cd85014b9e fix(plan-canvas): route session state through CLAUDE_CONFIG_DIR
The Plan Canvas library wrote ~/.claude/plan-canvas while the SessionStart
and Stop hooks read CLAUDE_CONFIG_DIR/plan-canvas. Under a second profile
SessionStart missed open reviews and Stop let a turn end with feedback
still queued.

Expand a leading ~ in CLAUDE_CONFIG_DIR the way scripts/lib/agent-data-home.js
does, so a profile written as ~/.claude-work resolves to one directory rather
than two.

Add hook-level coverage that spawns each hook with a distinct
CLAUDE_CONFIG_DIR and no ECC_* override, plus a producer-to-consumer round
trip. Eight of the nine cases fail against the previous revision.
2026-10-03 20:20:40 +05:30
Suliman Abdulrazzaq 238309ce29 Merge branch 'main' into fix/gateguard-session-budget 2026-10-03 10:29:38 +03:00
basil-k-aji-dev 00d73fbae2 fix(hooks): honor CLAUDE_CONFIG_DIR instead of writing to ~/.claude
Claude Code points every profile-local path at CLAUDE_CONFIG_DIR when it is
set, but ECC hooks kept reading and writing ~/.claude. A session launched
under a second profile left its bash log, cost tracker, metrics, session data
and skill runs in the default profile, where the other one reads them back.

Two groups, as reported:

- getDefaultClaudeAgentDataHome() in scripts/lib/agent-data-home.js now
  follows the variable, which covers every hook reaching ~/.claude through
  resolveAgentDataHome(). ECC_AGENT_DATA_HOME still wins, so the documented
  workaround keeps working.
- plan-canvas-pending, plan-canvas-sessions, mcp-health-check and
  post-bash-command-log hard-coded os.homedir()/.claude and now resolve the
  same way, matching scripts/hooks/ecc-statusline.js:68.

A blank value is ignored rather than resolved, since path.resolve('  ')
returns the process cwd and would scatter profile data through whatever
directory the session started in.

Fixes #3266
2026-10-03 12:34:14 +05:30
Affaan Mustafa c05b2d6614 chore: retire the Everything Claude Code name, ECC only (#3330)
* chore: retire the Everything Claude Code name, ECC only

Replace the expanded product name and the everything-claude-code slug
with ECC across READMEs, translated docs, harness folders, skills,
commands, scripts and examples. Repo URLs now point at affaan-m/ECC.

Rename repo-internal artifacts to ecc forms (ecc-conventions skill,
ecc-guardrails, ecc-research-playbook, ecc-team-config, ecc-instincts
with ecc-* ids) and update .claude/ecc-tools.json.

Legacy install ids, plugin dirs, package names and MCP prefixes stay
where existing installs resolve against them. The codex global-state
check and harness-audit repo detection accept the new name first and
the legacy one as a fallback. The Discord bot prefers ~/GitHub/ECC and
falls back to the nested legacy checkout path.

Refresh the skill-triggers registryDigest for the changed skill text.

* fix(ecc-tools): drop managed paths to the removed Claude repo skill

.claude/skills/<repo-skill>/SKILL.md was deleted in #2770 but stayed listed in
ecc-tools.json and the team config. Remove those entries and point the Claude
Code adapter at the existing .agents/skills/ecc-conventions/SKILL.md.
2026-10-01 17:50:41 -05:00
Affaan Mustafa 0348d7b6d7 chore(release): 2.2.3, ECC-only naming in pi/core and a longer npm publish wait (#3329)
pi/core and the bundled blueprint skill now refer to the project as ECC only;
the generated README no longer expands the name. Profile contents, package
name and paths are unchanged (123 skills, 24 commands).

The release workflows poll npm for up to five minutes after publish before
verifying the artifact. In 2.2.2 the check gave up after about 30 seconds while
npm was still processing, failing the run before latest was promoted and the
GitHub Release was created.

Version metadata synced to 2.2.3 with CHANGELOG, translated release headings
and release notes; skill-triggers registryDigest updated for the blueprint text.
2026-10-01 17:49:51 -05:00
phungthien269 8fca8ef4aa fix: address CodeRabbit and Greptile review findings
- Reject missing values for --ecc-root/--out with exit code 2 (CodeRabbit) | - Preserve regex matcher tokens with metacharacters or mcp__ prefixes from lower-casing (CodeRabbit) | - Shell-quote eccRoot in the env prefix to close CWE-78 command substitution (CodeRabbit, Greptile P1) | - Fix last_verified_at future date in the deepseek-harness record (CodeRabbit) | - Guide and skill: clarify skill-name placeholders and make the compose check verify the entry instead of echoing OK on exit status alone (Greptile P2 x2) | - Tests: 11 passing including new coverage for quoting, regex preservation, and flag validation
2026-10-02 00:49:31 +07:00
phungthien269 3756908464 feat: register DeepSeek Harness in the adapter compliance matrix
- Add deepseek-harness record (Adapter-backed) to ADAPTER_RECORDS with onramp, verification commands, risk notes, and source docs | - Regenerate the matrix block in docs/architecture/harness-adapter-compliance.md via the renderer | - Addresses the Harness-compatibility reference gap flagged by the ecc-tools Reference Set Readiness check on PR #3309
2026-10-02 00:26:01 +07:00
phungthien269 85dd488892 feat: add DeepSeek Harness guide, hooks generator, and setup skill
- docs/DEEPSEEK-HARNESS-GUIDE.md: skills roots, hooks bridge mounting, loader rules learned from practice, scope table | - scripts/dsh/generate-hooks-config.js: convert hooks/hooks.json to the dsh-hooks-claude-code bridge format (lower-case tool matchers, inline plugin-root env prefix, timeout preserved, unsupported events skipped) | - skills/deepseek-harness-setup: skills-first entry point per repo policy | - tests/scripts/dsh-generate-hooks-config.test.js: 8 unit + CLI tests
2026-10-02 00:17:22 +07:00
Yann Roberto 2d2712b899 fix(context): honor directive order and singular rejections 2026-10-01 15:03:34 +08:00
Yann Roberto 184c657788 fix(sessions): preserve committed results on lock cleanup errors 2026-10-01 14:59:29 +08:00
Jeffrey Montoya ab45ed1edf test(gateguard): pin questions for classes that skip profiling 2026-10-01 02:15:54 -04:00
Yann Roberto b1ba495997 fix(sessions): serialize alias read-modify-write operations 2026-10-01 13:27:55 +08:00
Yann Roberto f302863d58 fix(context): respect indirect skill citations in auto routing 2026-10-01 13:27:51 +08:00
Jeffrey Montoya b821262997 feat(gateguard): compare hook latency with confidence intervals 2026-09-30 20:27:36 -04:00
haelyra 3c17f41a43 docs: resolve integration review findings 2026-09-30 18:58:26 -04:00
haelyra 7c03203dd8 Merge remote-tracking branch 'refs/remotes/pr/3267' into maint/stewardship-batch-2026-09-30 2026-09-30 13:21:01 -04:00
haelyra cc9d2c3b65 Merge remote-tracking branch 'refs/remotes/pr/3255' into maint/stewardship-batch-2026-09-30 2026-09-30 13:21:01 -04:00
haelyra c965a61da5 Merge remote-tracking branch 'refs/remotes/pr/2655' into maint/stewardship-batch-2026-09-30
# Conflicts:
#	.claude-plugin/marketplace.json
#	.claude-plugin/plugin.json
#	AGENTS.md
#	README.md
#	README.zh-CN.md
#	docs/de-DE/README.md
#	docs/es/README.md
#	docs/ja-JP/README.md
#	docs/ko-KR/README.md
#	docs/pt-BR/README.md
#	docs/ru/README.md
#	docs/th/README.md
#	docs/tr/AGENTS.md
#	docs/tr/README.md
#	docs/uk-UA/README.md
#	docs/ur/README.md
#	docs/vi-VN/README.md
#	docs/zh-CN/AGENTS.md
#	docs/zh-CN/README.md
#	docs/zh-TW/README.md
#	scripts/lib/install-manifests.js
2026-09-30 13:20:41 -04:00
Yann Roberto 3768c03653 fix(state-store): preserve permission errors after lock retries 2026-10-01 01:03:41 +08:00
Yann Roberto 94cd7db7c3 fix(state-store): retry Windows contention and repair fixtures 2026-10-01 00:25:22 +08:00
Jeffrey Montoya 2ec52af6d0 perf(gateguard): profile a change only when used and check its window first 2026-09-30 12:13:18 -04:00
Jeffrey Montoya d39fcd8ac3 refactor(gateguard): read the transcript tail through a small shared module 2026-09-30 12:13:10 -04:00
Jeffrey Montoya d95aa3425e feat(gateguard): measure hook latency in fresh processes 2026-09-30 11:35:49 -04:00
Jeffrey Montoya eab92ed21d perf(gateguard): skip plain character runs when lexing code 2026-09-30 11:35:49 -04:00
Jeffrey Montoya b7f80ff012 perf(gateguard): load modules and parse searches only when a call needs them 2026-09-30 11:35:49 -04:00
drakeo338andClaude Sonnet 5.5 334b7847bd fix(hooks): keep ~/.claude eligible as project sharing root with CLAUDE_CONFIG_DIR
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-09-30 12:43:02 +00:00
Geronimo 5f7ba84b38 Merge branch 'affaan-m:main' into main 2026-09-30 16:02:28 +05:30
Jeffrey Montoya b15c0538f6 refactor(gateguard): split change profile and search filters into focused modules 2026-09-30 00:23:08 -04:00
Jeffrey Montoya 3055eb3746 fix(gateguard): link SARIF findings to the evaluated corpus 2026-09-30 00:04:11 -04:00
Jeffrey Montoya 1221a8727f fix(gateguard): honour directory-qualified include globs 2026-09-30 00:04:11 -04:00
drakeo338 01c9a670de fix(hooks): honor CLAUDE_CONFIG_DIR for the default agent data home
getDefaultClaudeAgentDataHome ignored CLAUDE_CONFIG_DIR and always returned ~/.claude, so sessions with a relocated Claude config dir wrote ECC data to the wrong place when ECC_AGENT_DATA_HOME was unset. Fixes #3266.
2026-09-30 03:06:16 +00:00
Jeffrey Montoya 216fe89c36 feat(gateguard): write scenario gate results as SARIF 2026-09-29 22:20:41 -04:00
Jeffrey Montoya 10639a0c9e fix(gateguard): gate NotebookEdit and hard-linked targets
NotebookEdit is gated as a first-touch edit of notebook_path.
A file with more than one hard link gets no credit, trivial, sibling or cap allowance.
2026-09-29 21:50:26 -04:00
Jeffrey Montoya 20eb2395e9 fix(gateguard): gate MultiEdit by its file_path and widen surface checks
Also refuses rg -z as read-only and never writes metrics through a symlink.
2026-09-29 21:50:26 -04:00
Jeffrey Montoya 6e677b008c fix(gateguard): judge comment-only edits against the file they change 2026-09-29 21:50:25 -04:00
Jeffrey Montoya 82420f061a fix(gateguard): treat directive comments and continuations as code 2026-09-29 21:50:24 -04:00
Jeffrey Montoya 42e9072136 feat(gateguard): credit test files by their module stem
Drop the hardcoded baseline alias from the evaluation script and refresh the results.
2026-09-29 21:50:23 -04:00
Jeffrey Montoya 21ea5f96a8 feat(gateguard): profile shell scripts and Python module constants 2026-09-29 21:50:22 -04:00