backprop — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited backprop (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.
Plan-then-execute fixes the code and forgets. SDD fixes the code AND edits spec so recurrence is impossible. That edit is backprop.
/sdd:build verification./sdd:check flags VIOLATE with root cause found.Read failure output / bug report. Find exact file:line of wrong behavior. Name root cause in one math-glyph sentence.
Ask three questions:
Draft the spec change. Never skip §B; §V/§I/§T are case-by-case.
Template:
§B row: B<next>|<date>|<root cause>|V<N>
§V line: V<next>: <testable rule that would have caught it>Example:
§B row: B<n>|<date>|refund job ran twice on retry|V<N>
§V line: V<N>: every refund → idempotency key check before charge reversalNew invariant without test = lie. Add failing test first. Name test so it cites the invariant: TestV<N>_RefundIdempotent.
Fix code. Run test. Must pass. Run full suite. Must not regress.
Commit spec edit + test + code fix together. Commit msg: backprop §B.<n> + §V.<N>: <one-line cause>.
Walks the six steps end-to-end on one bug. Numbers fictional; demonstrates judgment.
Input: bug: refund job double-charged customer on retry
refund(ctx, amount) not check prior charge state. §B row: B<n>|2026-04-20|refund retry double-charged because no idempotency check|V<N>
§V line: V<N>: every refund ! idempotency key check before charge reversalTestV<N>_RefundIdempotent — invoke refund twice w/ same key → expect ≤ 1 reversal posted.charge.reverse(). Run new test → pass. Run full suite → not regression.backprop §B.<n>(+) + §V.<N>(+): refund retry double-charge.! hold over not forbid).Bad: V<N>: code should be correct. Good: V<N>: every pg_query ! params interpolated via driver, not string concat.
i++ vs i-- in throwaway).Still append §B entry — record that this failure mode was considered. Future bug in same class → §B search shows precedent.
Every backprop run produces:
No dashboards. No log files. SPEC.md + git is the full history.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.