open/config-audit
1
0
Fork 0
Code Issues Releases 1 Activity

fix(permissions): param-aware DIS dead-allow + CNF conflict matching

Browse source 
The DIS scanner collapsed Tool(param) rules to the bare tool name, so
Agent(model:opus) deny + Agent(model:sonnet) allow (and the same for
WebFetch(domain:...)) were flagged as dead config — a false positive now
that CC 2.1.178 matches Tool(param:value) and 2.1.172 adds domain rules.
The conflict-detector shared the blind spot from the other side: a
wildcard deny like WebFetch(domain:*) did not cover a
WebFetch(domain:good.com) allow, so a genuine cross-scope conflict was
missed (false negative).

New shared scanners/lib/permission-rules.mjs:
- parseRule / paramMatches (glob)
- dominates(deny, allow) -> DIS dead-allow (deny fully covers allow)
- rulesIntersect(a, b)   -> CNF cross-scope conflict (match sets intersect)

DIS now delegates to dominates; conflict-detector :156 delegates to
rulesIntersect. A bare deny still covers all params, so true positives
are preserved (Bash deny + Bash(npm:*) allow still flagged).

Re-seeded the marketplace-medium snapshots: the false-positive CA-DIS
finding (Read(src/**) allow + Read(./.env) deny) is correctly gone. This
changes snapshot CONTENT only — envelope schema is unchanged, so --json
and --raw stay byte-stable.

Full suite: 837/837 green (+25). self-audit PASS, A(100)/A(97).
This commit is contained in:
Kjell Tore Guttormsen 2026-06-18 13:06:20 +02:00
commit bec3f45329
16 changed files with 324 additions and 145 deletions
Download patch file Download diff file Expand all files Collapse all files

2
tests/snapshots/default-output/posture.json
View file

@ -1,4 +1,4 @@
{
"kind": "text",
"payload": "`[CML] CLAUDE.md Linter`: 1 finding(s) (0ms)\n `[SET] Settings Validator`: 0 finding(s) (0ms)\n `[HKV] Hook Validator`: 0 finding(s) (0ms)\n `[RUL] Rules Validator`: 0 finding(s) (0ms)\n `[MCP] MCP Config Validator`: 0 finding(s) (0ms)\n `[IMP] Import Resolver`: 0 finding(s) (0ms)\n `[CNF] Conflict Detector`: 0 finding(s) (0ms)\n `[GAP] Feature Gap Scanner`: 17 finding(s) (0ms)\n `[TOK] Token Hotspots`: 1 finding(s) (0ms)\n `[CPS] Cache-Prefix Stability`: 0 finding(s) (0ms)\n `[DIS] Disabled-In-Schema`: 1 finding(s) (0ms)\n `[COL] Plugin Skill Collision`: 0 finding(s) (0ms)\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n Configuration health\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\n Health: A (97/100) — Healthy setup, only minor polish needed\n 9 areas reviewed\n\n Area scores\n ───────────\n `CLAUDE.md` ........... A (90) `Settings` ............ A (90)\n `Hooks` ............... A (100) `Rules` ............... A (100)\n `MCP` ................. A (100) `Imports` ............. A (100)\n `Conflicts` ........... A (100) `Token Efficiency` .... A (90)\n `Plugin Hygiene` ...... A (100)\n\n 17 ways you could get more out of Claude Code — see /config-audit feature-gap\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"
"payload": "`[CML] CLAUDE.md Linter`: 1 finding(s) (0ms)\n `[SET] Settings Validator`: 0 finding(s) (0ms)\n `[HKV] Hook Validator`: 0 finding(s) (0ms)\n `[RUL] Rules Validator`: 0 finding(s) (0ms)\n `[MCP] MCP Config Validator`: 0 finding(s) (0ms)\n `[IMP] Import Resolver`: 0 finding(s) (0ms)\n `[CNF] Conflict Detector`: 0 finding(s) (0ms)\n `[GAP] Feature Gap Scanner`: 17 finding(s) (0ms)\n `[TOK] Token Hotspots`: 1 finding(s) (0ms)\n `[CPS] Cache-Prefix Stability`: 0 finding(s) (0ms)\n `[DIS] Disabled-In-Schema`: 0 finding(s) (0ms)\n `[COL] Plugin Skill Collision`: 0 finding(s) (0ms)\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n Configuration health\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\n Health: A (98/100) — Healthy setup, only minor polish needed\n 9 areas reviewed\n\n Area scores\n ───────────\n `CLAUDE.md` ........... A (90) `Settings` ............ A (100)\n `Hooks` ............... A (100) `Rules` ............... A (100)\n `MCP` ................. A (100) `Imports` ............. A (100)\n `Conflicts` ........... A (100) `Token Efficiency` .... A (90)\n `Plugin Hygiene` ...... A (100)\n\n 17 ways you could get more out of Claude Code — see /config-audit feature-gap\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"
}

27
tests/snapshots/default-output/scan-orchestrator.json
View file

@ -517,29 +517,12 @@
"status": "ok",
"files_scanned": 1,
"duration_ms": 0,
"findings": [
{
"id": "CA-DIS-001",
"scanner": "DIS",
"severity": "low",
"title": "A tool is in both the let-in list and the shut-out list",
"description": "When a tool is in both lists, the shut-out always wins, so the let-in entry does nothing. It looks like the tool is approved, but it isn't.",
"file": "/Users/ktg/repos/ktg-plugin-marketplace/config-audit/tests/fixtures/marketplace-medium/.claude/settings.json",
"line": null,
"evidence": "Read: allow=\"Read(src/**)\" + deny=\"Read(./.env)\"",
"category": "permissions-hygiene",
"recommendation": "Decide whether the tool should be allowed or denied, and remove it from the other list.",
"autoFixable": false,
"userImpactCategory": "Dead config",
"userActionLanguage": "Optional cleanup",
"relevanceContext": "test-fixture-no-impact"
}
],
"findings": [],
"counts": {
"critical": 0,
"high": 0,
"medium": 0,
"low": 1,
"low": 0,
"info": 0
}
},
@ -559,15 +542,15 @@
}
],
"aggregate": {
"total_findings": 20,
"total_findings": 19,
"counts": {
"critical": 0,
"high": 0,
"medium": 1,
"low": 8,
"low": 7,
"info": 11
},
"risk_score": 12,
"risk_score": 11,
"risk_band": "Medium",
"verdict": "PASS",
"scanners_ok": 11,

10
tests/snapshots/v5.0.0-stderr/posture.txt
View file

@ -1,4 +1,4 @@
[CML] CLAUDE.md Linter: 1 finding(s) (13ms)
[CML] CLAUDE.md Linter: 1 finding(s) (14ms)
[SET] Settings Validator: 0 finding(s) (1ms)
[HKV] Hook Validator: 0 finding(s) (1ms)
[RUL] Rules Validator: 0 finding(s) (1ms)
@ -8,18 +8,18 @@
[GAP] Feature Gap Scanner: 17 finding(s) (2ms)
[TOK] Token Hotspots: 1 finding(s) (8ms)
[CPS] Cache-Prefix Stability: 0 finding(s) (1ms)
[DIS] Disabled-In-Schema: 1 finding(s) (1ms)
[COL] Plugin Skill Collision: 0 finding(s) (0ms)
[DIS] Disabled-In-Schema: 0 finding(s) (0ms)
[COL] Plugin Skill Collision: 0 finding(s) (1ms)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Config-Audit Health Score
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Health: A (97/100) 9 areas scanned
Health: A (98/100) 9 areas scanned
Area Scores
───────────
CLAUDE.md ........... A (90) Settings ............ A (90)
CLAUDE.md ........... A (90) Settings ............ A (100)
Hooks ............... A (100) Rules ............... A (100)
MCP ................. A (100) Imports ............. A (100)
Conflicts ........... A (100) Token Efficiency .... A (90)

25
tests/snapshots/v5.0.0/drift.json
View file

@ -248,29 +248,16 @@
"category": "token-efficiency",
"recommendation": "Install the package locally (so detect-mcp-tool-count can read its manifest), or run the server once and cache its tools/list response under ~/.claude/config-audit/mcp-cache/<name>.json. See knowledge/cache-telemetry-recipe.md.",
"autoFixable": false
},
{
"id": "CA-DIS-001",
"scanner": "DIS",
"severity": "low",
"title": "Tool listed in both permissions.deny and permissions.allow",
"description": ".claude/settings.json contains 1 tool present in both deny and allow lists. The deny list wins — the allow entries are dead config but still load on every turn and may confuse future readers about intent.",
"file": "/Users/ktg/repos/ktg-plugin-marketplace/config-audit/tests/fixtures/marketplace-medium/.claude/settings.json",
"line": null,
"evidence": "Read: allow=\"Read(src/**)\" + deny=\"Read(./.env)\"",
"category": "permissions-hygiene",
"recommendation": "Remove the redundant allow entries. If you actually want this tool enabled, remove it from the deny list instead. Settings should express intent clearly.",
"autoFixable": false
}
],
"movedFindings": [],
"scoreChange": {
"before": {
"score": 91,
"score": 92,
"grade": "A"
},
"after": {
"score": 91,
"score": 92,
"grade": "A"
},
"delta": 0
@ -291,11 +278,11 @@
{
"name": "Settings",
"before": {
"score": 90,
"score": 100,
"grade": "A"
},
"after": {
"score": 90,
"score": 100,
"grade": "A"
},
"delta": 0
@ -398,8 +385,8 @@
}
],
"summary": {
"totalBefore": 20,
"totalAfter": 20,
"totalBefore": 19,
"totalAfter": 19,
"newCount": 0,
"resolvedCount": 0,
"trend": "stable"

6
tests/snapshots/v5.0.0/fix-cli.json
View file

@ -118,12 +118,6 @@
"title": "High MCP tool-schema budget on server \"memory\"",
"file": ".mcp.json",
"recommendation": "Install the package locally (so detect-mcp-tool-count can read its manifest), or run the server once and cache its tools/list response under ~/.claude/config-audit/mcp-cache/<name>.json. See knowledge/cache-telemetry-recipe.md."
},
{
"findingId": "CA-DIS-001",
"title": "Tool listed in both permissions.deny and permissions.allow",
"file": "/Users/ktg/repos/ktg-plugin-marketplace/config-audit/tests/fixtures/marketplace-medium/.claude/settings.json",
"recommendation": "Remove the redundant allow entries. If you actually want this tool enabled, remove it from the deny list instead. Settings should express intent clearly."
}
],
"backupId": null

36
tests/snapshots/v5.0.0/posture.json
View file

@ -24,8 +24,8 @@
"id": "settings",
"name": "Settings",
"grade": "A",
"score": 90,
"findingCount": 1
"score": 100,
"findingCount": 0
},
{
"id": "hooks",
@ -94,7 +94,7 @@
"scannerEnvelope": {
"meta": {
"target": "/Users/ktg/repos/ktg-plugin-marketplace/config-audit/tests/fixtures/marketplace-medium",
"timestamp": "2026-06-18T10:18:35.634Z",
"timestamp": "2026-06-18T11:03:14.339Z",
"version": "2.2.0",
"tool": "config-audit"
},
@ -103,7 +103,7 @@
"scanner": "CML",
"status": "ok",
"files_scanned": 1,
"duration_ms": 3,
"duration_ms": 2,
"findings": [
{
"id": "CA-CML-001",
@ -187,7 +187,7 @@
"scanner": "IMP",
"status": "ok",
"files_scanned": 1,
"duration_ms": 2,
"duration_ms": 1,
"findings": [],
"counts": {
"critical": 0,
@ -551,26 +551,12 @@
"status": "ok",
"files_scanned": 1,
"duration_ms": 0,
"findings": [
{
"id": "CA-DIS-001",
"scanner": "DIS",
"severity": "low",
"title": "Tool listed in both permissions.deny and permissions.allow",
"description": ".claude/settings.json contains 1 tool present in both deny and allow lists. The deny list wins — the allow entries are dead config but still load on every turn and may confuse future readers about intent.",
"file": "/Users/ktg/repos/ktg-plugin-marketplace/config-audit/tests/fixtures/marketplace-medium/.claude/settings.json",
"line": null,
"evidence": "Read: allow=\"Read(src/**)\" + deny=\"Read(./.env)\"",
"category": "permissions-hygiene",
"recommendation": "Remove the redundant allow entries. If you actually want this tool enabled, remove it from the deny list instead. Settings should express intent clearly.",
"autoFixable": false
}
],
"findings": [],
"counts": {
"critical": 0,
"high": 0,
"medium": 0,
"low": 1,
"low": 0,
"info": 0
}
},
@ -578,7 +564,7 @@
"scanner": "COL",
"status": "ok",
"files_scanned": 0,
"duration_ms": 0,
"duration_ms": 1,
"findings": [],
"counts": {
"critical": 0,
@ -590,15 +576,15 @@
}
],
"aggregate": {
"total_findings": 20,
"total_findings": 19,
"counts": {
"critical": 0,
"high": 0,
"medium": 1,
"low": 8,
"low": 7,
"info": 11
},
"risk_score": 12,
"risk_score": 11,
"risk_band": "Medium",
"verdict": "PASS",
"scanners_ok": 11,

34
tests/snapshots/v5.0.0/scan-orchestrator.json
View file

@ -1,7 +1,7 @@
{
"meta": {
"target": "/Users/ktg/repos/ktg-plugin-marketplace/config-audit/tests/fixtures/marketplace-medium",
"timestamp": "2026-06-18T10:18:35.439Z",
"timestamp": "2026-06-18T11:03:14.159Z",
"version": "2.2.0",
"tool": "config-audit"
},
@ -122,7 +122,7 @@
"scanner": "GAP",
"status": "ok",
"files_scanned": 4,
"duration_ms": 2,
"duration_ms": 1,
"findings": [
{
"id": "CA-GAP-001",
@ -358,7 +358,7 @@
"scanner": "TOK",
"status": "ok",
"files_scanned": 2,
"duration_ms": 8,
"duration_ms": 7,
"findings": [
{
"id": "CA-TOK-001",
@ -457,27 +457,13 @@
"scanner": "DIS",
"status": "ok",
"files_scanned": 1,
"duration_ms": 1,
"findings": [
{
"id": "CA-DIS-001",
"scanner": "DIS",
"severity": "low",
"title": "Tool listed in both permissions.deny and permissions.allow",
"description": ".claude/settings.json contains 1 tool present in both deny and allow lists. The deny list wins — the allow entries are dead config but still load on every turn and may confuse future readers about intent.",
"file": "/Users/ktg/repos/ktg-plugin-marketplace/config-audit/tests/fixtures/marketplace-medium/.claude/settings.json",
"line": null,
"evidence": "Read: allow=\"Read(src/**)\" + deny=\"Read(./.env)\"",
"category": "permissions-hygiene",
"recommendation": "Remove the redundant allow entries. If you actually want this tool enabled, remove it from the deny list instead. Settings should express intent clearly.",
"autoFixable": false
}
],
"duration_ms": 0,
"findings": [],
"counts": {
"critical": 0,
"high": 0,
"medium": 0,
"low": 1,
"low": 0,
"info": 0
}
},
@ -485,7 +471,7 @@
"scanner": "COL",
"status": "ok",
"files_scanned": 0,
"duration_ms": 0,
"duration_ms": 1,
"findings": [],
"counts": {
"critical": 0,
@ -497,15 +483,15 @@
}
],
"aggregate": {
"total_findings": 20,
"total_findings": 19,
"counts": {
"critical": 0,
"high": 0,
"medium": 1,
"low": 8,
"low": 7,
"info": 11
},
"risk_score": 12,
"risk_score": 11,
"risk_band": "Medium",
"verdict": "PASS",
"scanners_ok": 11,