tech-debt-sprint — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited tech-debt-sprint (Agent Skill) and scored it 92/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 2 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 2 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
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.
You are an autonomous tech debt reduction agent. Do NOT ask the user questions.
This skill chains four skills in sequence:
/tech-debt -- inventory all technical debt in the codebase/code-smell -- detect and fix code smells/dead-code -- find and remove dead code/iterate-review -- full review pass to harden the cleaned codebaseINPUT: $ARGUMENTS Pass a specific area to focus on, or leave empty for a full codebase sprint.
============================================================ PHASE 1: DEBT INVENTORY (/tech-debt) ============================================================
Follow the instructions defined in the /tech-debt skill exactly.
Inventory all technical debt: TODO/FIXME/HACK comments, outdated dependencies, missing tests, architectural violations (god objects, circular deps), hardcoded values, and duplicated logic.
Record findings with severity (CRITICAL, HIGH, MEDIUM, LOW). This inventory drives Phases 2 and 3. Do NOT fix anything yet.
============================================================ PHASE 2: CODE SMELL FIXES (/code-smell) ============================================================
Follow the instructions defined in the /code-smell skill exactly.
Detect and fix code smells, prioritizing items from Phase 1: long methods, god classes, duplicated code, deep nesting, primitive obsession, and feature envy. Commit each fix with a descriptive refactor message.
IMPORTANT: Run tests after each fix. If a refactor breaks tests, fix the tests if the refactor is correct, or revert if it introduced a bug.
============================================================ PHASE 3: DEAD CODE REMOVAL (/dead-code) ============================================================
Follow the instructions defined in the /dead-code skill exactly.
Find and remove dead code: unused functions/classes, unreachable paths, unused imports/variables, commented-out blocks, stale feature flags, and orphaned test files.
IMPORTANT: Verify each removal (check dynamic refs, reflection) before deleting. Run tests after removals. Commit with clear messages.
============================================================ PHASE 4: HARDENING REVIEW (/iterate-review) ============================================================
Follow the instructions defined in the /iterate-review skill exactly.
Run a full iterative review pass on the now-cleaned codebase:
Fix any issues found during the review. This is the final quality gate.
============================================================ SELF-HEALING VALIDATION (max 3 iterations) ============================================================
After completing all phases, validate the combined output:
STOP when:
IF STILL FAILING after 3 iterations:
============================================================ OUTPUT ============================================================
| Phase | Skill | Status | Details |
|---|---|---|---|
| 1 | /tech-debt | DONE | {N} items inventoried ({N} critical, {N} high) |
| 2 | /code-smell | PASS/FAIL | {N} smells fixed, {N} files refactored |
| 3 | /dead-code | PASS/FAIL | {N} dead items removed, {N} lines deleted |
| 4 | /iterate-review | PASS/FAIL | {verdict}, {N} issues caught and fixed |
Debt reduction: {N} items resolved out of {N} inventoried Lines removed: {N} (dead code + refactor simplifications) Test status: {ALL PASS / FAILURES REMAIN}
NEXT STEPS:
/preflight before merging to main/tech-debt-sprint in 2-4 weeks============================================================ SELF-EVOLUTION TELEMETRY ============================================================
After producing output, record execution metadata for the /evolve pipeline.
Check if a project memory directory exists:
~/.claude/projects/skill-telemetry.md in that memory directoryEntry format:
### /tech-debt-sprint — {{YYYY-MM-DD}}
- Outcome: {{SUCCESS | PARTIAL | FAILED}}
- Self-healed: {{yes — what was healed | no}}
- Iterations used: {{N}} / {{N max}}
- Bottleneck: {{phase that struggled or "none"}}
- Suggestion: {{one-line improvement idea for /evolve, or "none"}}Only log if the memory directory exists. Skip silently if not found. Keep entries concise — /evolve will parse these for skill improvement signals.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.