open/config-audit
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity Actions
config-audit/tests/scanners
History
Kjell Tore Guttormsen 2a3cb537f9 feat(scanner): add AGT per-agent description bloat advisory (v5.9 B1-rest) 
Flag any single active agent whose description exceeds a 500-char soft cap
(the same bloat threshold TOK pattern F applies to SKILL.md descriptions).
Every char of an agent description re-enters context in the always-loaded
agent listing on every turn, so a long one is a per-turn cost.

Framing is deliberately ADVISORY, severity low: unlike CA-SKL-001 agents
have NO verified per-description cap, so nothing is dropped — this is a bloat
advisory, never a truncation claim. Per-agent advisories are emitted BEFORE
the aggregate roll-up (mirrors SKL 001->002).

- PER_AGENT_DESC_SOFT_CAP (=500) added to lib/agent-listing-budget.mjs as the
  single source of truth, disclosed as a heuristic.
- Per-agent loop reuses measureActiveAgentListing()'s agents[] (name, source,
  pluginName, path, descLength) — no new enumeration.
- 5 new tests (fire/strict-boundary/no-truncation-claim/recommendation/order).

Byte-stable: AGT already stripped from frozen v5.0.0 baselines; SC-5 unchanged
(HOME-scoped, hermetic HOME has no agents). Suite green: 1181 tests.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-23 16:27:24 +02:00
..
accurate-tokens.test.mjs test: isolate HOME in all CLI-spawning tests (close leak class) 2026-06-18 18:17:09 +02:00
agent-listing-scanner.test.mjs feat(scanner): add AGT per-agent description bloat advisory (v5.9 B1-rest) 2026-06-23 16:27:24 +02:00
cache-prefix.test.mjs feat(config-audit): cache-prefix stability scanner CPS (v5 N3) [skip-docs] 2026-05-01 07:37:54 +02:00
campaign-cli.test.mjs feat(campaign): cross-repo prioritized backlog (v5.7 Fase 2 Block 4b) 2026-06-23 02:44:21 +02:00
campaign-export-cli.test.mjs feat(campaign): plan export + execution-by-reuse (v5.7 Fase 2 Block 4c) 2026-06-23 10:08:04 +02:00
campaign-write-cli.test.mjs feat(campaign): /config-audit campaign — human-approved write-CLI + command (v5.7 Fase 2 Block 3c) 2026-06-22 17:25:15 +02:00
claude-md-linter.test.mjs feat(rul,cml): durability findings — config lost after compaction (v5.5.0 / A) 2026-06-20 11:20:04 +02:00
cli-humanizer.test.mjs feat(tok,acr): v5.6 B2 — load-pattern column in token-hotspots 2026-06-20 20:21:01 +02:00
collision.test.mjs feat(config-audit): cross-plugin collision scanner COL (v5 N6) [skip-docs] 2026-05-01 07:46:15 +02:00
conflict-detector.test.mjs fix(permissions): param-aware DIS dead-allow + CNF conflict matching 2026-06-18 13:06:20 +02:00
disabled-in-schema.test.mjs feat(dis): flag forbidden-param permission rules CC silently ignores 2026-06-19 14:04:25 +02:00
drift-cli.test.mjs test: isolate HOME in all CLI-spawning tests (close leak class) 2026-06-18 18:17:09 +02:00
feature-gap-scanner.test.mjs feat(feature-gap): recommend disableBundledSkills under skill-listing pressure 2026-06-18 21:38:19 +02:00
fix-cli.test.mjs test(fix-cli): close missed HOME-leak — isolate spawns + lock SKL/COL out 2026-06-18 18:50:20 +02:00
fix-engine.test.mjs feat(ultraplan-local): v1.6.0 — /ultraresearch-local deep research command 2026-04-08 08:58:35 +02:00
hook-validator.test.mjs fix(hkv): remove post-session — it is a runner hook, not a settings event 2026-06-20 10:12:52 +02:00
import-resolver.test.mjs feat(ultraplan-local): v1.6.0 — /ultraresearch-local deep research command 2026-04-08 08:58:35 +02:00
knowledge-refresh-cli.test.mjs feat(knowledge): knowledge-refresh — living register, deterministic stale core + web poll (v5.7 Chunk 3) 2026-06-21 19:22:57 +02:00
lint-default-output.test.mjs test(snapshots): make byte/snapshot tests hermetic + re-seed baseline 2026-06-18 12:26:00 +02:00
manifest.test.mjs feat(mft): v5.6 B1 — load-pattern accounting in manifest 2026-06-20 17:48:55 +02:00
mcp-config-validator.test.mjs fix(mcp-config-validator): remove invented trust field (verify-first) 2026-06-18 14:22:56 +02:00
optimization-lens-scanner.test.mjs feat(opt): optimization lens CA-OPT-001 (procedure→skill) — v5.7 Fase 1 Chunk 2a 2026-06-20 23:10:03 +02:00
output-style-scanner.test.mjs feat(ost): v5.6 C — output-style scanner (CA-OST, count 13→14) 2026-06-20 21:02:44 +02:00
plugin-health-scanner.test.mjs feat(plh): flag plugin-agent frontmatter Claude Code ignores (v5.5.0 / E) 2026-06-20 11:11:12 +02:00
posture-grade-stability.test.mjs feat(skill-listing): add CA-SKL-002 aggregate listing-budget check 2026-06-18 18:06:17 +02:00
posture-humanizer.test.mjs feat(ost): v5.6 C — output-style scanner (CA-OST, count 13→14) 2026-06-20 21:02:44 +02:00
posture.test.mjs test: isolate HOME in all CLI-spawning tests (close leak class) 2026-06-18 18:17:09 +02:00
rollback-engine.test.mjs feat(ultraplan-local): v1.6.0 — /ultraresearch-local deep research command 2026-04-08 08:58:35 +02:00
rules-validator.test.mjs feat(acr,yaml): v5.6 Foundation — load-pattern enumeration + block-seq parser 2026-06-20 16:57:52 +02:00
scan-orchestrator-humanizer.test.mjs feat(scanner): add AGT agent-listing always-loaded budget finding (v5.9 B1) 2026-06-23 15:02:59 +02:00
scan-orchestrator.test.mjs feat(ultraplan-local): v1.6.0 — /ultraresearch-local deep research command 2026-04-08 08:58:35 +02:00
self-audit.test.mjs feat(config-audit): self-audit --check-readme flag (v5 F6) [skip-docs] 2026-05-01 07:09:26 +02:00
settings-validator.test.mjs feat(set): validate autoMode structure + flag it in shared settings (CA-SET) 2026-06-19 22:14:26 +02:00
skill-listing-scanner.test.mjs feat(skill-listing): add CA-SKL-002 aggregate listing-budget check 2026-06-18 18:06:17 +02:00
token-hotspots-cli.test.mjs test: isolate HOME in all CLI-spawning tests (close leak class) 2026-06-18 18:17:09 +02:00
token-hotspots.test.mjs test(tok): hermetic HOME in runScanner — deterministic on populated ~/.claude 2026-06-21 13:45:53 +02:00