Commit Graph

5 Commits

Author SHA1 Message Date
Moseyuh333
9899a6dacc fix(github-skills): add missing import, add unit tests, fix settings JSON parse (#6186)
feat(skills): GitHub skill-discovery subsystem — search/score/scan/import agent skills from GitHub, MCP tools (scope-gated) + /api/github-skills route (host-pinned api.github.com, sanitized errors). Merged — thank you, @Moseyuh333! Pre-merge fixes: passed toolDef.scopes so tools gate correctly under scope enforcement, routed the GET error path through sanitizeErrorMessage+500 (Hard Rule #12), and realigned the agent-skills count guards (catalog/routes/generator/mcp) for the new catalog entry. Base-reds only. Integrated into release/v3.8.45.
2026-07-05 08:28:01 -03:00
diegosouzapw
3ea416350e fix(tests+ci): update 42→43 skill count, fix E2E artifact path
- Unit/integration tests: update hardcoded 42→43 in 7 test files
  (agentSkillTools-mcp, agentSkills-catalog, agentSkills-generator,
  agent-skills-content, agent-skills-discovery, listCapabilities-a2a)
  to match the 43rd skill (config-codex-cli) added in the previous commit.
- Include CONFIG_SKILL_IDS in integration content test ALL_IDS so
  skills/config-codex-cli/ is no longer "unexpected".
- listCapabilities.ts: change totalSkills from literal 42 to catalog.length
  so it adapts to catalog growth automatically.
- computeCoverage assertions: include config.have in totalSkills check.
- CI: switch E2E artifact from upload-artifact path (ambiguous stripping)
  to explicit tar archive. Fixes "Could not find a production build in
  ./.build/next" — the previous approach's download path was double-nested
  (.build/next/next/...) due to upload-artifact LCA computation. tar -czf
  stores .build/next/... relative to CWD; tar -xzf restores them verbatim.
- Also exclude .build/next/cache from the tar to keep archive lean.
- feat(translator): strip client_metadata in Responses→Chat translation
  (Mistral 422 extra_forbidden fix); add regression test.
2026-06-08 10:41:00 -03:00
diegosouzapw
8f0615fd04 fix(security): resolve 5 CodeQL alerts, document FPs, harden deploy skills
Real fixes (alerts auto-close on next scan):
- agentSkills/generator: escape backslash before double-quote in YAML
  frontmatter so a trailing backslash can't escape the closing quote
  (js/incomplete-sanitization #297/#298)
- usage: replace /\s*\(RESTRICTED\)\s*$/ with a non-backtracking literal +
  trim (js/polynomial-redos #275)
- i18n/request: skip __proto__/constructor/prototype in deepMergeFallback
  (js/prototype-pollution-utility #274)
- scripts/ad-hoc/nvidia diag: log key presence only, never key chars
  (js/clear-text-logging #273)
- tests: regression coverage for all three production fixes (YAML round-trip,
  proto-pollution guard, RESTRICTED strip + ReDoS timing)

Docs:
- ARCHITECTURE.md: executor count 45->55, OAuth modules 15->16, agy.ts in list

Deploy skills (deploy-vps-*-cc/-cx/-ag): add --legacy-peer-deps (npm v11
peer-dep resolver crashes on the omniroute tree) and replace the
"; pm2 start" that masked a failed install with a proper && chain.
2026-06-01 19:36:48 -03:00
diegosouzapw
c73a90134b feat(agent-skills): rewrite generator templates to English (GAP-D - template change) 2026-05-28 13:55:32 -03:00
diegosouzapw
d1d0ddda0f test(agent-skills): unit tests for generator + scripts smoke
20 tests covering:
- dry-run produces report without writing (filesystem unchanged)
- apply writes SKILL.md for API and CLI skills with correct sections
- apply writes all 42 skills when no filter
- idempotency: second run reports 0 generated, all unchanged
- prune dry-run detects orphans without deleting
- prune apply deletes orphan dirs, preserves catalog dirs
- marker preservation: custom block survives regeneration
- buildSkillMarkdown: valid shape, no escape errors, correct sections per category
- onlyIds filter limits generation
- report shape validation

Also adds gray-matter as dependency (per plan prerequisite check).
2026-05-27 21:55:01 -03:00