@ryuenn3123/agentic-senior-core 2.0.5 → 2.0.7
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/.agent-context/blueprints/mobile-app.md +91 -21
- package/.agent-context/profiles/platform.md +13 -13
- package/.agent-context/profiles/regulated.md +13 -13
- package/.agent-context/profiles/startup.md +13 -13
- package/.agent-context/review-checklists/frontend-skill-parity.md +28 -28
- package/.agent-context/review-checklists/frontend-usability.md +33 -33
- package/.agent-context/review-checklists/release-operations.md +29 -29
- package/.agent-context/skills/README.md +62 -62
- package/.agent-context/skills/backend/README.md +67 -67
- package/.agent-context/skills/backend/architecture.md +360 -360
- package/.agent-context/skills/backend/compatibility-manifest.json +8 -8
- package/.agent-context/skills/backend/data-access.md +230 -230
- package/.agent-context/skills/backend/errors.md +137 -137
- package/.agent-context/skills/backend/validation.md +116 -116
- package/.agent-context/skills/backend.md +28 -28
- package/.agent-context/skills/cli/README.md +55 -49
- package/.agent-context/skills/cli/compatibility-manifest.json +8 -8
- package/.agent-context/skills/cli/init.md +37 -37
- package/.agent-context/skills/cli/output.md +35 -35
- package/.agent-context/skills/cli/safety-telemetry.md +39 -0
- package/.agent-context/skills/cli/upgrade.md +37 -37
- package/.agent-context/skills/cli.md +31 -28
- package/.agent-context/skills/distribution/.evidence/compatibility-manifest.json +9 -0
- package/.agent-context/skills/distribution/.evidence/sbom-excerpt.json +6 -0
- package/.agent-context/skills/distribution/.evidence/test-report.json +8 -0
- package/.agent-context/skills/distribution/CHANGELOG.md +7 -0
- package/.agent-context/skills/distribution/README.md +27 -19
- package/.agent-context/skills/distribution/compatibility-manifest.json +8 -8
- package/.agent-context/skills/distribution/compatibility.md +31 -31
- package/.agent-context/skills/distribution/package.json +5 -0
- package/.agent-context/skills/distribution/provenance-attestation.md +47 -0
- package/.agent-context/skills/distribution/publish.md +36 -36
- package/.agent-context/skills/distribution/rollback.md +31 -31
- package/.agent-context/skills/distribution/tests/.gitkeep +1 -0
- package/.agent-context/skills/distribution.md +31 -28
- package/.agent-context/skills/frontend/.evidence/compatibility-manifest.json +9 -0
- package/.agent-context/skills/frontend/.evidence/sbom-excerpt.json +6 -0
- package/.agent-context/skills/frontend/.evidence/test-report.json +8 -0
- package/.agent-context/skills/frontend/CHANGELOG.md +7 -0
- package/.agent-context/skills/frontend/README.md +49 -36
- package/.agent-context/skills/frontend/accessibility.md +107 -107
- package/.agent-context/skills/frontend/compatibility-manifest.json +8 -8
- package/.agent-context/skills/frontend/conversion-clarity.md +51 -0
- package/.agent-context/skills/frontend/motion.md +66 -66
- package/.agent-context/skills/frontend/package.json +5 -0
- package/.agent-context/skills/frontend/performance.md +62 -62
- package/.agent-context/skills/frontend/responsive-delivery.md +41 -0
- package/.agent-context/skills/frontend/tests/.gitkeep +1 -0
- package/.agent-context/skills/frontend/ui-architecture.md +128 -128
- package/.agent-context/skills/frontend.md +35 -29
- package/.agent-context/skills/fullstack/.evidence/compatibility-manifest.json +9 -0
- package/.agent-context/skills/fullstack/.evidence/sbom-excerpt.json +6 -0
- package/.agent-context/skills/fullstack/.evidence/test-report.json +8 -0
- package/.agent-context/skills/fullstack/CHANGELOG.md +7 -0
- package/.agent-context/skills/fullstack/README.md +27 -19
- package/.agent-context/skills/fullstack/compatibility-manifest.json +8 -8
- package/.agent-context/skills/fullstack/contracts.md +52 -52
- package/.agent-context/skills/fullstack/end-to-end.md +41 -41
- package/.agent-context/skills/fullstack/feature-slicing.md +64 -64
- package/.agent-context/skills/fullstack/package.json +5 -0
- package/.agent-context/skills/fullstack/release-coordination.md +51 -0
- package/.agent-context/skills/fullstack/tests/.gitkeep +1 -0
- package/.agent-context/skills/fullstack.md +29 -26
- package/.agent-context/skills/index.json +107 -107
- package/.agent-context/skills/review-quality/.evidence/compatibility-manifest.json +9 -0
- package/.agent-context/skills/review-quality/.evidence/sbom-excerpt.json +6 -0
- package/.agent-context/skills/review-quality/.evidence/test-report.json +8 -0
- package/.agent-context/skills/review-quality/CHANGELOG.md +7 -0
- package/.agent-context/skills/review-quality/README.md +27 -19
- package/.agent-context/skills/review-quality/benchmark.md +29 -29
- package/.agent-context/skills/review-quality/compatibility-manifest.json +8 -8
- package/.agent-context/skills/review-quality/package.json +5 -0
- package/.agent-context/skills/review-quality/planning.md +37 -37
- package/.agent-context/skills/review-quality/release-decision.md +49 -0
- package/.agent-context/skills/review-quality/security.md +33 -33
- package/.agent-context/skills/review-quality/tests/.gitkeep +1 -0
- package/.agent-context/skills/review-quality.md +30 -27
- package/.agent-context/stacks/flutter.md +16 -16
- package/.agent-context/stacks/react-native.md +16 -16
- package/.agent-context/state/architecture-map.md +25 -25
- package/.agent-context/state/benchmark-analysis.json +431 -431
- package/.agent-context/state/benchmark-thresholds.json +10 -10
- package/.agent-context/state/benchmark-watchlist.json +19 -19
- package/.agent-context/state/dependency-map.md +32 -32
- package/.agent-context/state/quality-trend-report.json +16 -6
- package/.agent-context/state/skill-platform.json +38 -38
- package/.agent-context/state/weekly-governance-report.json +126 -0
- package/.agent-override.md +36 -36
- package/.cursorrules +1 -1
- package/.gemini/instructions.md +20 -20
- package/.github/ISSUE_TEMPLATE/v1.7-frontend-work-item.yml +54 -54
- package/.github/copilot-instructions.md +20 -20
- package/.github/workflows/benchmark-detection.yml +38 -38
- package/.github/workflows/benchmark-intelligence.yml +50 -50
- package/.github/workflows/frontend-usability-gate.yml +36 -36
- package/.github/workflows/governance-weekly-report.yml +43 -0
- package/.github/workflows/release-gate.yml +32 -32
- package/.github/workflows/sbom-compliance.yml +32 -32
- package/.windsurfrules +1 -1
- package/AGENTS.md +27 -27
- package/README.md +383 -368
- package/lib/cli/commands/optimize.mjs +171 -171
- package/lib/cli/compatibility.mjs +124 -124
- package/lib/cli/constants.mjs +35 -0
- package/lib/cli/token-optimization.mjs +275 -275
- package/lib/cli/utils.mjs +4 -1
- package/mcp.json +92 -92
- package/package.json +2 -1
- package/scripts/benchmark-gate.mjs +121 -121
- package/scripts/benchmark-intelligence.mjs +140 -140
- package/scripts/detection-benchmark.mjs +138 -138
- package/scripts/frontend-usability-audit.mjs +87 -87
- package/scripts/generate-sbom.mjs +61 -61
- package/scripts/governance-weekly-report.mjs +293 -0
- package/scripts/init-project.ps1 +104 -104
- package/scripts/llm-judge.mjs +664 -664
- package/scripts/quality-trend-report.mjs +288 -288
- package/scripts/release-gate.mjs +261 -259
- package/scripts/skill-tier-policy.mjs +75 -75
- package/scripts/token-optimization-benchmark.mjs +252 -252
- package/scripts/validate.mjs +874 -865
|
@@ -1,107 +1,107 @@
|
|
|
1
|
-
{
|
|
2
|
-
"defaultTier": "advance",
|
|
3
|
-
"tiers": [
|
|
4
|
-
{
|
|
5
|
-
"name": "standard",
|
|
6
|
-
"description": "Compatibility fallback only"
|
|
7
|
-
},
|
|
8
|
-
{
|
|
9
|
-
"name": "advance",
|
|
10
|
-
"description": "Default operating tier"
|
|
11
|
-
},
|
|
12
|
-
{
|
|
13
|
-
"name": "expert",
|
|
14
|
-
"description": "Complex architecture and integration"
|
|
15
|
-
},
|
|
16
|
-
{
|
|
17
|
-
"name": "above",
|
|
18
|
-
"description": "Release-critical enterprise governance"
|
|
19
|
-
}
|
|
20
|
-
],
|
|
21
|
-
"domains": {
|
|
22
|
-
"frontend": {
|
|
23
|
-
"name": "frontend",
|
|
24
|
-
"displayName": "Frontend",
|
|
25
|
-
"description": "Unified frontend delivery covering UI architecture, motion, accessibility, and conversion clarity.",
|
|
26
|
-
"defaultTier": "advance",
|
|
27
|
-
"defaultPackFileName": "frontend.md",
|
|
28
|
-
"evidence": "Frontend usability audit, accessibility checks, and visual regression output.",
|
|
29
|
-
"tierToPackFileNames": {
|
|
30
|
-
"standard": "frontend.md",
|
|
31
|
-
"advance": "frontend.md",
|
|
32
|
-
"expert": "frontend.md",
|
|
33
|
-
"above": "frontend.md"
|
|
34
|
-
}
|
|
35
|
-
},
|
|
36
|
-
"backend": {
|
|
37
|
-
"name": "backend",
|
|
38
|
-
"displayName": "Backend",
|
|
39
|
-
"description": "Backend delivery with strict layer separation, validation, and safe data access.",
|
|
40
|
-
"defaultTier": "advance",
|
|
41
|
-
"defaultPackFileName": "backend.md",
|
|
42
|
-
"evidence": "Unit tests, API contracts, and release gate output.",
|
|
43
|
-
"tierToPackFileNames": {
|
|
44
|
-
"standard": "backend.md",
|
|
45
|
-
"advance": "backend.md",
|
|
46
|
-
"expert": "backend.md",
|
|
47
|
-
"above": "backend.md"
|
|
48
|
-
}
|
|
49
|
-
},
|
|
50
|
-
"fullstack": {
|
|
51
|
-
"name": "fullstack",
|
|
52
|
-
"displayName": "Fullstack",
|
|
53
|
-
"description": "Single-path product delivery across frontend and backend boundaries.",
|
|
54
|
-
"defaultTier": "advance",
|
|
55
|
-
"defaultPackFileName": "fullstack.md",
|
|
56
|
-
"evidence": "End-to-end tests, contract validation, and feature parity review.",
|
|
57
|
-
"tierToPackFileNames": {
|
|
58
|
-
"standard": "fullstack.md",
|
|
59
|
-
"advance": "fullstack.md",
|
|
60
|
-
"expert": "fullstack.md",
|
|
61
|
-
"above": "fullstack.md"
|
|
62
|
-
}
|
|
63
|
-
},
|
|
64
|
-
"cli": {
|
|
65
|
-
"name": "cli",
|
|
66
|
-
"displayName": "CLI",
|
|
67
|
-
"description": "Smart command-line delivery with safe defaults, machine-readable output, and upgrade flows.",
|
|
68
|
-
"defaultTier": "advance",
|
|
69
|
-
"defaultPackFileName": "cli.md",
|
|
70
|
-
"evidence": "CLI smoke tests, dry-run output, and automation-friendly reports.",
|
|
71
|
-
"tierToPackFileNames": {
|
|
72
|
-
"standard": "cli.md",
|
|
73
|
-
"advance": "cli.md",
|
|
74
|
-
"expert": "cli.md",
|
|
75
|
-
"above": "cli.md"
|
|
76
|
-
}
|
|
77
|
-
},
|
|
78
|
-
"distribution": {
|
|
79
|
-
"name": "distribution",
|
|
80
|
-
"displayName": "Distribution",
|
|
81
|
-
"description": "Package safety, rollback, compatibility, and release hygiene.",
|
|
82
|
-
"defaultTier": "expert",
|
|
83
|
-
"defaultPackFileName": "distribution.md",
|
|
84
|
-
"evidence": "Install validation, rollback verification, and publish dry-run logs.",
|
|
85
|
-
"tierToPackFileNames": {
|
|
86
|
-
"standard": "distribution.md",
|
|
87
|
-
"advance": "distribution.md",
|
|
88
|
-
"expert": "distribution.md",
|
|
89
|
-
"above": "distribution.md"
|
|
90
|
-
}
|
|
91
|
-
},
|
|
92
|
-
"review-quality": {
|
|
93
|
-
"name": "review-quality",
|
|
94
|
-
"displayName": "Review Quality",
|
|
95
|
-
"description": "Planning, review, benchmark, and security discipline for every change.",
|
|
96
|
-
"defaultTier": "expert",
|
|
97
|
-
"defaultPackFileName": "review-quality.md",
|
|
98
|
-
"evidence": "PR review output, benchmark report, and security audit results.",
|
|
99
|
-
"tierToPackFileNames": {
|
|
100
|
-
"standard": "review-quality.md",
|
|
101
|
-
"advance": "review-quality.md",
|
|
102
|
-
"expert": "review-quality.md",
|
|
103
|
-
"above": "review-quality.md"
|
|
104
|
-
}
|
|
105
|
-
}
|
|
106
|
-
}
|
|
107
|
-
}
|
|
1
|
+
{
|
|
2
|
+
"defaultTier": "advance",
|
|
3
|
+
"tiers": [
|
|
4
|
+
{
|
|
5
|
+
"name": "standard",
|
|
6
|
+
"description": "Compatibility fallback only"
|
|
7
|
+
},
|
|
8
|
+
{
|
|
9
|
+
"name": "advance",
|
|
10
|
+
"description": "Default operating tier"
|
|
11
|
+
},
|
|
12
|
+
{
|
|
13
|
+
"name": "expert",
|
|
14
|
+
"description": "Complex architecture and integration"
|
|
15
|
+
},
|
|
16
|
+
{
|
|
17
|
+
"name": "above",
|
|
18
|
+
"description": "Release-critical enterprise governance"
|
|
19
|
+
}
|
|
20
|
+
],
|
|
21
|
+
"domains": {
|
|
22
|
+
"frontend": {
|
|
23
|
+
"name": "frontend",
|
|
24
|
+
"displayName": "Frontend",
|
|
25
|
+
"description": "Unified frontend delivery covering UI architecture, motion, accessibility, and conversion clarity.",
|
|
26
|
+
"defaultTier": "advance",
|
|
27
|
+
"defaultPackFileName": "frontend.md",
|
|
28
|
+
"evidence": "Frontend usability audit, accessibility checks, and visual regression output.",
|
|
29
|
+
"tierToPackFileNames": {
|
|
30
|
+
"standard": "frontend.md",
|
|
31
|
+
"advance": "frontend.md",
|
|
32
|
+
"expert": "frontend.md",
|
|
33
|
+
"above": "frontend.md"
|
|
34
|
+
}
|
|
35
|
+
},
|
|
36
|
+
"backend": {
|
|
37
|
+
"name": "backend",
|
|
38
|
+
"displayName": "Backend",
|
|
39
|
+
"description": "Backend delivery with strict layer separation, validation, and safe data access.",
|
|
40
|
+
"defaultTier": "advance",
|
|
41
|
+
"defaultPackFileName": "backend.md",
|
|
42
|
+
"evidence": "Unit tests, API contracts, and release gate output.",
|
|
43
|
+
"tierToPackFileNames": {
|
|
44
|
+
"standard": "backend.md",
|
|
45
|
+
"advance": "backend.md",
|
|
46
|
+
"expert": "backend.md",
|
|
47
|
+
"above": "backend.md"
|
|
48
|
+
}
|
|
49
|
+
},
|
|
50
|
+
"fullstack": {
|
|
51
|
+
"name": "fullstack",
|
|
52
|
+
"displayName": "Fullstack",
|
|
53
|
+
"description": "Single-path product delivery across frontend and backend boundaries.",
|
|
54
|
+
"defaultTier": "advance",
|
|
55
|
+
"defaultPackFileName": "fullstack.md",
|
|
56
|
+
"evidence": "End-to-end tests, contract validation, and feature parity review.",
|
|
57
|
+
"tierToPackFileNames": {
|
|
58
|
+
"standard": "fullstack.md",
|
|
59
|
+
"advance": "fullstack.md",
|
|
60
|
+
"expert": "fullstack.md",
|
|
61
|
+
"above": "fullstack.md"
|
|
62
|
+
}
|
|
63
|
+
},
|
|
64
|
+
"cli": {
|
|
65
|
+
"name": "cli",
|
|
66
|
+
"displayName": "CLI",
|
|
67
|
+
"description": "Smart command-line delivery with safe defaults, machine-readable output, and upgrade flows.",
|
|
68
|
+
"defaultTier": "advance",
|
|
69
|
+
"defaultPackFileName": "cli.md",
|
|
70
|
+
"evidence": "CLI smoke tests, dry-run output, and automation-friendly reports.",
|
|
71
|
+
"tierToPackFileNames": {
|
|
72
|
+
"standard": "cli.md",
|
|
73
|
+
"advance": "cli.md",
|
|
74
|
+
"expert": "cli.md",
|
|
75
|
+
"above": "cli.md"
|
|
76
|
+
}
|
|
77
|
+
},
|
|
78
|
+
"distribution": {
|
|
79
|
+
"name": "distribution",
|
|
80
|
+
"displayName": "Distribution",
|
|
81
|
+
"description": "Package safety, rollback, compatibility, and release hygiene.",
|
|
82
|
+
"defaultTier": "expert",
|
|
83
|
+
"defaultPackFileName": "distribution.md",
|
|
84
|
+
"evidence": "Install validation, rollback verification, and publish dry-run logs.",
|
|
85
|
+
"tierToPackFileNames": {
|
|
86
|
+
"standard": "distribution.md",
|
|
87
|
+
"advance": "distribution.md",
|
|
88
|
+
"expert": "distribution.md",
|
|
89
|
+
"above": "distribution.md"
|
|
90
|
+
}
|
|
91
|
+
},
|
|
92
|
+
"review-quality": {
|
|
93
|
+
"name": "review-quality",
|
|
94
|
+
"displayName": "Review Quality",
|
|
95
|
+
"description": "Planning, review, benchmark, and security discipline for every change.",
|
|
96
|
+
"defaultTier": "expert",
|
|
97
|
+
"defaultPackFileName": "review-quality.md",
|
|
98
|
+
"evidence": "PR review output, benchmark report, and security audit results.",
|
|
99
|
+
"tierToPackFileNames": {
|
|
100
|
+
"standard": "review-quality.md",
|
|
101
|
+
"advance": "review-quality.md",
|
|
102
|
+
"expert": "review-quality.md",
|
|
103
|
+
"above": "review-quality.md"
|
|
104
|
+
}
|
|
105
|
+
}
|
|
106
|
+
}
|
|
107
|
+
}
|
|
@@ -0,0 +1,9 @@
|
|
|
1
|
+
{
|
|
2
|
+
"schemaVersion": "compatibility-manifest-v1",
|
|
3
|
+
"artifactType": "skill-domain-evidence",
|
|
4
|
+
"domain": "review-quality",
|
|
5
|
+
"ides": ["cursor", "windsurf", "copilot", "gemini", "claude", "codex", "cline"],
|
|
6
|
+
"nodeMin": "18",
|
|
7
|
+
"platforms": ["windows", "linux", "macos"],
|
|
8
|
+
"validatedAt": "2026-04-11T12:00:00Z"
|
|
9
|
+
}
|
|
@@ -1,19 +1,27 @@
|
|
|
1
|
-
# Review Quality Skills
|
|
2
|
-
|
|
3
|
-
Default tier: `expert`
|
|
4
|
-
|
|
5
|
-
This domain formalizes review quality across planning discipline, security enforcement, and benchmark-driven decision making.
|
|
6
|
-
|
|
7
|
-
## Topics
|
|
8
|
-
- [Planning](planning.md) - Plan quality, scope control, and change strategy
|
|
9
|
-
- [Security](security.md) - Critical vulnerability policy and boundary safeguards
|
|
10
|
-
- [Benchmarking](benchmark.md) - Regression detection and evidence-based comparison
|
|
11
|
-
|
|
12
|
-
|
|
13
|
-
|
|
14
|
-
-
|
|
15
|
-
|
|
16
|
-
|
|
17
|
-
-
|
|
18
|
-
-
|
|
19
|
-
-
|
|
1
|
+
# Review Quality Skills
|
|
2
|
+
|
|
3
|
+
Default tier: `expert`
|
|
4
|
+
|
|
5
|
+
This domain formalizes review quality across planning discipline, security enforcement, and benchmark-driven decision making.
|
|
6
|
+
|
|
7
|
+
## Topics
|
|
8
|
+
- [Planning](planning.md) - Plan quality, scope control, and change strategy
|
|
9
|
+
- [Security](security.md) - Critical vulnerability policy and boundary safeguards
|
|
10
|
+
- [Benchmarking](benchmark.md) - Regression detection and evidence-based comparison
|
|
11
|
+
- [Release Decisioning](release-decision.md) - Explicit readiness verdicts, blocker ownership, and escalation logic
|
|
12
|
+
|
|
13
|
+
## Operating Model
|
|
14
|
+
- Use `expert` for standard review workflows.
|
|
15
|
+
- Escalate to `above` for release-critical or governance-sensitive changes.
|
|
16
|
+
|
|
17
|
+
## Above-Line Additions
|
|
18
|
+
- Security halt protocol for critical findings.
|
|
19
|
+
- Benchmark gate thresholds integrated in CI.
|
|
20
|
+
- Review evidence bundle for auditability.
|
|
21
|
+
|
|
22
|
+
## Usage Example
|
|
23
|
+
|
|
24
|
+
```bash
|
|
25
|
+
node ./scripts/governance-weekly-report.mjs
|
|
26
|
+
node ./scripts/release-gate.mjs
|
|
27
|
+
```
|
|
@@ -1,30 +1,30 @@
|
|
|
1
|
-
# Benchmarking
|
|
2
|
-
|
|
3
|
-
Tier: ADVANCE
|
|
4
|
-
|
|
5
|
-
Benchmarking should guide release decisions with measurable evidence rather than anecdotal impressions.
|
|
6
|
-
|
|
7
|
-
## Benchmark Inputs
|
|
8
|
-
|
|
9
|
-
- Baseline benchmark snapshot.
|
|
10
|
-
- Current candidate benchmark output.
|
|
11
|
-
- Threshold policy for acceptable deltas.
|
|
12
|
-
|
|
13
|
-
## Evaluation Rules
|
|
14
|
-
|
|
15
|
-
- Treat regressions above threshold as release blockers.
|
|
16
|
-
- Require owner assignment for all negative deltas.
|
|
17
|
-
- Re-run benchmark after remediation to verify recovery.
|
|
18
|
-
|
|
19
|
-
## Reporting Standard
|
|
20
|
-
|
|
21
|
-
- Include benchmark summary in release bundle.
|
|
22
|
-
- Keep machine-readable JSON output for automation.
|
|
23
|
-
- Track trend over time, not only single-run status.
|
|
24
|
-
|
|
25
|
-
## Review Checklist
|
|
26
|
-
|
|
27
|
-
- [ ] Baseline and current benchmark are comparable.
|
|
28
|
-
- [ ] Threshold policy is applied consistently.
|
|
29
|
-
- [ ] Regressions have owner and due date.
|
|
1
|
+
# Benchmarking
|
|
2
|
+
|
|
3
|
+
Tier: ADVANCE
|
|
4
|
+
|
|
5
|
+
Benchmarking should guide release decisions with measurable evidence rather than anecdotal impressions.
|
|
6
|
+
|
|
7
|
+
## Benchmark Inputs
|
|
8
|
+
|
|
9
|
+
- Baseline benchmark snapshot.
|
|
10
|
+
- Current candidate benchmark output.
|
|
11
|
+
- Threshold policy for acceptable deltas.
|
|
12
|
+
|
|
13
|
+
## Evaluation Rules
|
|
14
|
+
|
|
15
|
+
- Treat regressions above threshold as release blockers.
|
|
16
|
+
- Require owner assignment for all negative deltas.
|
|
17
|
+
- Re-run benchmark after remediation to verify recovery.
|
|
18
|
+
|
|
19
|
+
## Reporting Standard
|
|
20
|
+
|
|
21
|
+
- Include benchmark summary in release bundle.
|
|
22
|
+
- Keep machine-readable JSON output for automation.
|
|
23
|
+
- Track trend over time, not only single-run status.
|
|
24
|
+
|
|
25
|
+
## Review Checklist
|
|
26
|
+
|
|
27
|
+
- [ ] Baseline and current benchmark are comparable.
|
|
28
|
+
- [ ] Threshold policy is applied consistently.
|
|
29
|
+
- [ ] Regressions have owner and due date.
|
|
30
30
|
- [ ] Reports are archived for audit trail.
|
|
@@ -1,8 +1,8 @@
|
|
|
1
|
-
{
|
|
2
|
-
"schemaVersion": "compatibility-manifest-v1",
|
|
3
|
-
"artifactType": "skill-domain",
|
|
4
|
-
"domain": "review-quality",
|
|
5
|
-
"ides": ["cursor", "windsurf", "copilot", "gemini", "claude", "codex", "cline"],
|
|
6
|
-
"nodeMin": "18",
|
|
7
|
-
"platforms": ["windows", "linux", "macos"]
|
|
8
|
-
}
|
|
1
|
+
{
|
|
2
|
+
"schemaVersion": "compatibility-manifest-v1",
|
|
3
|
+
"artifactType": "skill-domain",
|
|
4
|
+
"domain": "review-quality",
|
|
5
|
+
"ides": ["cursor", "windsurf", "copilot", "gemini", "claude", "codex", "cline"],
|
|
6
|
+
"nodeMin": "18",
|
|
7
|
+
"platforms": ["windows", "linux", "macos"]
|
|
8
|
+
}
|
|
@@ -1,38 +1,38 @@
|
|
|
1
|
-
# Planning
|
|
2
|
-
|
|
3
|
-
Tier: EXPERT
|
|
4
|
-
|
|
5
|
-
Planning quality determines implementation quality. Non-trivial work should start with a scoped, testable plan.
|
|
6
|
-
|
|
7
|
-
## Planning Standard
|
|
8
|
-
|
|
9
|
-
- Define objective, scope, and non-goals.
|
|
10
|
-
- Identify architecture boundaries affected.
|
|
11
|
-
- Document risk, rollback, and validation strategy.
|
|
12
|
-
- Break work into reviewable increments.
|
|
13
|
-
|
|
14
|
-
## Required Plan Artifacts
|
|
15
|
-
|
|
16
|
-
- Implementation steps with expected output.
|
|
17
|
-
- File impact map.
|
|
18
|
-
- Test strategy and acceptance criteria.
|
|
19
|
-
- Migration notes if behavior changes.
|
|
20
|
-
|
|
21
|
-
## Scope Control
|
|
22
|
-
|
|
23
|
-
- Avoid unrelated refactors in feature commits.
|
|
24
|
-
- Split high-risk work into isolated checkpoints.
|
|
25
|
-
- Preserve public API compatibility unless explicitly versioned.
|
|
26
|
-
|
|
27
|
-
## Evidence Discipline
|
|
28
|
-
|
|
29
|
-
- Attach command output for validation and test evidence.
|
|
30
|
-
- Keep assumption logs explicit so reviewers can challenge unknowns early.
|
|
31
|
-
- Ensure rollback path is documented before implementation starts.
|
|
32
|
-
|
|
33
|
-
## Review Checklist
|
|
34
|
-
|
|
35
|
-
- [ ] Plan includes explicit success criteria.
|
|
36
|
-
- [ ] Risks and mitigations are documented.
|
|
37
|
-
- [ ] Validation commands are defined.
|
|
1
|
+
# Planning
|
|
2
|
+
|
|
3
|
+
Tier: EXPERT
|
|
4
|
+
|
|
5
|
+
Planning quality determines implementation quality. Non-trivial work should start with a scoped, testable plan.
|
|
6
|
+
|
|
7
|
+
## Planning Standard
|
|
8
|
+
|
|
9
|
+
- Define objective, scope, and non-goals.
|
|
10
|
+
- Identify architecture boundaries affected.
|
|
11
|
+
- Document risk, rollback, and validation strategy.
|
|
12
|
+
- Break work into reviewable increments.
|
|
13
|
+
|
|
14
|
+
## Required Plan Artifacts
|
|
15
|
+
|
|
16
|
+
- Implementation steps with expected output.
|
|
17
|
+
- File impact map.
|
|
18
|
+
- Test strategy and acceptance criteria.
|
|
19
|
+
- Migration notes if behavior changes.
|
|
20
|
+
|
|
21
|
+
## Scope Control
|
|
22
|
+
|
|
23
|
+
- Avoid unrelated refactors in feature commits.
|
|
24
|
+
- Split high-risk work into isolated checkpoints.
|
|
25
|
+
- Preserve public API compatibility unless explicitly versioned.
|
|
26
|
+
|
|
27
|
+
## Evidence Discipline
|
|
28
|
+
|
|
29
|
+
- Attach command output for validation and test evidence.
|
|
30
|
+
- Keep assumption logs explicit so reviewers can challenge unknowns early.
|
|
31
|
+
- Ensure rollback path is documented before implementation starts.
|
|
32
|
+
|
|
33
|
+
## Review Checklist
|
|
34
|
+
|
|
35
|
+
- [ ] Plan includes explicit success criteria.
|
|
36
|
+
- [ ] Risks and mitigations are documented.
|
|
37
|
+
- [ ] Validation commands are defined.
|
|
38
38
|
- [ ] Scope remains aligned with original objective.
|
|
@@ -0,0 +1,49 @@
|
|
|
1
|
+
# Release Decisioning
|
|
2
|
+
|
|
3
|
+
Tier: EXPERT
|
|
4
|
+
|
|
5
|
+
Release decisioning converts technical signals into explicit ship, hold, or rollback recommendations with ownership.
|
|
6
|
+
|
|
7
|
+
## Decision Inputs
|
|
8
|
+
|
|
9
|
+
Use a standardized input set for every release review:
|
|
10
|
+
|
|
11
|
+
- Validation and test status.
|
|
12
|
+
- Release gate and forbidden-content status.
|
|
13
|
+
- Benchmark and quality trend posture.
|
|
14
|
+
- Security and architecture findings.
|
|
15
|
+
- Trust-tier posture for required skill domains.
|
|
16
|
+
|
|
17
|
+
## Decision Outcomes
|
|
18
|
+
|
|
19
|
+
Each review must conclude with one outcome:
|
|
20
|
+
|
|
21
|
+
- Ship: all mandatory gates pass and no unresolved critical findings.
|
|
22
|
+
- Hold: one or more blockers remain unresolved.
|
|
23
|
+
- Rollback: post-release signal confirms unacceptable risk or regression.
|
|
24
|
+
|
|
25
|
+
## Blocker Policy
|
|
26
|
+
|
|
27
|
+
A blocker record needs:
|
|
28
|
+
|
|
29
|
+
- Title and category.
|
|
30
|
+
- Owner and deadline.
|
|
31
|
+
- User impact statement.
|
|
32
|
+
- Mitigation and validation command.
|
|
33
|
+
|
|
34
|
+
No blocker should remain in implicit or undocumented state.
|
|
35
|
+
|
|
36
|
+
## Escalation Rules
|
|
37
|
+
|
|
38
|
+
Escalate to maintainers immediately when:
|
|
39
|
+
|
|
40
|
+
- Critical security issue is detected.
|
|
41
|
+
- Gate output becomes inconsistent across environments.
|
|
42
|
+
- Rollback readiness cannot be proven.
|
|
43
|
+
|
|
44
|
+
## Review Checklist
|
|
45
|
+
|
|
46
|
+
- [ ] Decision outcome is explicit (ship/hold/rollback).
|
|
47
|
+
- [ ] Blockers include owner and due date.
|
|
48
|
+
- [ ] Validation evidence is attached to decision log.
|
|
49
|
+
- [ ] Escalation happened for critical unresolved risks.
|
|
@@ -1,34 +1,34 @@
|
|
|
1
|
-
# Security
|
|
2
|
-
|
|
3
|
-
Tier: EXPERT
|
|
4
|
-
|
|
5
|
-
Security review is a release gate. Critical vulnerabilities must halt feature rollout until resolved.
|
|
6
|
-
|
|
7
|
-
## Security Halt Policy
|
|
8
|
-
|
|
9
|
-
Immediately block release when any of the following are present:
|
|
10
|
-
- Hardcoded credentials or tokens.
|
|
11
|
-
- Injection vulnerabilities (SQL/command/template).
|
|
12
|
-
- Authentication or authorization bypass.
|
|
13
|
-
- Unvalidated external input entering privileged paths.
|
|
14
|
-
|
|
15
|
-
## Boundary Safeguards
|
|
16
|
-
|
|
17
|
-
- Validate external input at transport boundary.
|
|
18
|
-
- Enforce parameterized queries.
|
|
19
|
-
- Limit privilege scope for runtime credentials.
|
|
20
|
-
- Keep secrets in approved secret stores.
|
|
21
|
-
|
|
22
|
-
## Review Expectations
|
|
23
|
-
|
|
24
|
-
- Threat model relevant attack surfaces.
|
|
25
|
-
- Verify dependency vulnerability posture.
|
|
26
|
-
- Confirm error messages do not leak sensitive internals.
|
|
27
|
-
- Confirm audit logs exist for security-sensitive actions.
|
|
28
|
-
|
|
29
|
-
## Review Checklist
|
|
30
|
-
|
|
31
|
-
- [ ] No critical vulnerability remains open.
|
|
32
|
-
- [ ] Input and output boundaries are validated.
|
|
33
|
-
- [ ] Secret handling follows policy.
|
|
1
|
+
# Security
|
|
2
|
+
|
|
3
|
+
Tier: EXPERT
|
|
4
|
+
|
|
5
|
+
Security review is a release gate. Critical vulnerabilities must halt feature rollout until resolved.
|
|
6
|
+
|
|
7
|
+
## Security Halt Policy
|
|
8
|
+
|
|
9
|
+
Immediately block release when any of the following are present:
|
|
10
|
+
- Hardcoded credentials or tokens.
|
|
11
|
+
- Injection vulnerabilities (SQL/command/template).
|
|
12
|
+
- Authentication or authorization bypass.
|
|
13
|
+
- Unvalidated external input entering privileged paths.
|
|
14
|
+
|
|
15
|
+
## Boundary Safeguards
|
|
16
|
+
|
|
17
|
+
- Validate external input at transport boundary.
|
|
18
|
+
- Enforce parameterized queries.
|
|
19
|
+
- Limit privilege scope for runtime credentials.
|
|
20
|
+
- Keep secrets in approved secret stores.
|
|
21
|
+
|
|
22
|
+
## Review Expectations
|
|
23
|
+
|
|
24
|
+
- Threat model relevant attack surfaces.
|
|
25
|
+
- Verify dependency vulnerability posture.
|
|
26
|
+
- Confirm error messages do not leak sensitive internals.
|
|
27
|
+
- Confirm audit logs exist for security-sensitive actions.
|
|
28
|
+
|
|
29
|
+
## Review Checklist
|
|
30
|
+
|
|
31
|
+
- [ ] No critical vulnerability remains open.
|
|
32
|
+
- [ ] Input and output boundaries are validated.
|
|
33
|
+
- [ ] Secret handling follows policy.
|
|
34
34
|
- [ ] Least-privilege access is enforced.
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
# Review-quality skill test fixtures placeholder
|
|
@@ -1,28 +1,31 @@
|
|
|
1
|
-
# Review Quality Skill Pack
|
|
2
|
-
|
|
3
|
-
Default tier: `expert`
|
|
4
|
-
|
|
5
|
-
## Purpose
|
|
6
|
-
Turn code review, planning, and benchmark output into explicit quality decisions.
|
|
7
|
-
|
|
8
|
-
## In Scope
|
|
9
|
-
- Plan-first workflows
|
|
10
|
-
- Reviewer self-critique
|
|
11
|
-
- Benchmark interpretation
|
|
12
|
-
- Security and architecture audits
|
|
13
|
-
- Evidence-driven approval
|
|
14
|
-
|
|
15
|
-
|
|
16
|
-
-
|
|
17
|
-
- Every
|
|
18
|
-
-
|
|
19
|
-
-
|
|
20
|
-
|
|
21
|
-
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
-
|
|
25
|
-
-
|
|
26
|
-
|
|
27
|
-
|
|
1
|
+
# Review Quality Skill Pack
|
|
2
|
+
|
|
3
|
+
Default tier: `expert`
|
|
4
|
+
|
|
5
|
+
## Purpose
|
|
6
|
+
Turn code review, planning, and benchmark output into explicit quality decisions.
|
|
7
|
+
|
|
8
|
+
## In Scope
|
|
9
|
+
- Plan-first workflows
|
|
10
|
+
- Reviewer self-critique
|
|
11
|
+
- Benchmark interpretation
|
|
12
|
+
- Security and architecture audits
|
|
13
|
+
- Evidence-driven approval
|
|
14
|
+
- Explicit release decisions with blocker tracking and ownership
|
|
15
|
+
|
|
16
|
+
## Must-Have Checks
|
|
17
|
+
- Every rejection includes a reason and a fix
|
|
18
|
+
- Every approval includes evidence
|
|
19
|
+
- Benchmark deltas are explicit
|
|
20
|
+
- Security findings stop release when critical
|
|
21
|
+
- Release readiness verdict includes blockers, owner, and due date
|
|
22
|
+
|
|
23
|
+
## Evidence
|
|
24
|
+
- PR review report
|
|
25
|
+
- Security audit output
|
|
26
|
+
- Benchmark report
|
|
27
|
+
- Approval or rejection rationale
|
|
28
|
+
- Weekly governance report and release readiness summary
|
|
29
|
+
|
|
30
|
+
## Fallback
|
|
28
31
|
- Standard mode can be used only for low-risk maintenance and still requires written evidence.
|