complexity-gate — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited complexity-gate (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Register: DISCIPLINE — runs before the rest of the registry for software-development tasks. Goal: Choose how much rigor is required without weakening the core invariants. Core principle: compressed rigor is allowed; skipped rigor is not.
Announce: "Using complexity-gate to select the right Super Skills tier."
Score each input 0-3.
| Input | 0 | 1 | 2 | 3 |
|---|---|---|---|---|
| Integration points | local only | one module boundary | external service/API | auth/payment/PII/compliance boundary |
| Data sensitivity | none | non-sensitive app data | user/business-critical data | PII/secrets/payment/compliance |
| Irreversibility | no persisted effect | code-only revert | migration/backfill/config rollout | irreversible user/data/customer impact |
| Blast radius | one file/local | one module | user-facing service | cross-service/production-wide |
| Ambiguity | exact task | minor unknowns | multiple design choices | unclear intent/spec |
Total score routes the tier:
C0: 0-2 Direct/minimal
C1: 3-4 Bounded local change
C2: 5-7 Feature or behavior change
C3: 8-10 Production/multi-component
C4: 11+ Critical/high-assuranceIf any single input is 3, consider raising at least to C3 unless the scope is provably isolated.
Required:
Required:
tdd-verified);verification-before-completion before success claim;pattern-globalize if a bug was fixed.Required:
brainstorming lite;writing-plans lite;verification-before-completion;pattern-globalize for bug fixes.Required:
domain-alignment if project/domain is new;brainstorming;audit-design;writing-plans + task-risk-score;specialist-review;verification-before-completion;adr-commit;knowledge-compound.Required:
privacy-secrets-gate when data sensitivity is 2 or 3, or secrets/PII/payment/regulated data may appear;release-readiness before merge/deploy/traffic;| Tier | Spec requirement |
|---|---|
| C0 | No spec needed. Task description is sufficient. |
| C1 | Inline spec: 3-5 bullet points in task description. No lifecycle. |
| C2 | Lightweight spec: PROBLEM + SOLUTION + ACCEPTANCE CRITERIA. Single state: DRAFT → DONE. |
| C3 | Standard spec: DRAFT → APPROVED → LIVING-SPEC. Run brainstorming with full checklist including Step 7b (DPS SYSTEM INTENT). |
| C4 | Full lifecycle. Mandatory sign-off before each phase transition. All SYSTEM INTENT fields required. |
C0-C2 tasks should not use the full DPS spec lifecycle — proportional ceremony prevents adoption drag.
Never skip these:
No unverified completion claim.
No bug fix without checking bug class.
No architectural/contract decision without a record.
No repeated failure without systematic-debugging/spec-conflict stop.
No sensitive-data change without `privacy-secrets-gate` when data sensitivity is 2 or 3.
No real learning lost after a cycle.COMPLEXITY GATE
Scores: integration=<0-3>, data=<0-3>, irreversibility=<0-3>, blast=<0-3>, ambiguity=<0-3>
Tier: C0|C1|C2|C3|C4
Spec tier: none|inline|lightweight|standard|full
Required skills:
Optional skills:
Reasoning:
Evidence constraints:<!-- Populated by knowledge-compound after cycles where this skill underperformed --> <!-- New entries should follow shared/gotcha-schema.md -->
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.