Standards — focused engineering guidance

Load the smallest set of standards justified by the caller's files, language, and risks. Do not preload the entire reference corpus.

Prompt

Check standards for the changed files in fleet-router PR #214: cli/internal/auth/token.go and cli/internal/auth/token_test.go, Go, a security-sensitive change. Load only the matching references and report cited findings with path and line plus checked and not-checked scope.

It's working if

• The report loads common-standards.md plus only the matching Go reference, never the full reference corpus.
• Every finding cites a path and line, e.g. cli/internal/auth/token.go:18.
• The response discloses checked and not_checked scope explicitly, even when not_checked is empty.
• git diff --stat shows no test, gate, or fixture file changed; standards reports findings only.

Procedure

1. Record the supplied paths, language, change type, and risk cues.
2. Load common-standards.md plus only the matching language or checklist references.
3. Compare the supplied artifact to those sources.
4. Return cited findings with path and line when possible, plus checked and not-checked scope.
5. Stop.

This skill provides context and findings. It does not edit, validate, retry, approve, commit, release, deliver, or decide continuation.

Load-bearing conventions for produced code (MEASURED)

When the caller is about to WRITE code (not only review it), surface the matching language rules INLINE in the working context — a behind-the-link reference does not change behavior; an inline imperative does. The Go core:

• Wrap every propagated error with context: fmt.Errorf("doing X: %w", err) — never return a bare inner error.
• Multi-case functions get TABLE-DRIVEN tests ([]struct cases + t.Run per case), asserting exact expected values including the error cases.

For other languages, pull the matching reference below and inline its top rules the same way.

Measured 2026-08-04, probe standards-go-conventions (gpt-5.6-luna, N=2, directional): control produced the %w-wrapped + table-driven shape in 1/2 runs; with these rules inline, 2/2. Inline-imperative beats reference-link — the graphify probe measured a linked doc instruction obeyed 0/2. Ledger: evals/skill-probes/LEDGER.md.

Mutation-safety standards

When the supplied change rewrites existing files in bulk — formatters, codemods, migration scripts, generators pointed at hand-written sources — check it against three standards and report each as a finding when absent:

• Single audited mutation chokepoint. All rewrites flow through one named command or script whose inputs, outputs, and dry-run mode can be inspected. Edits scattered across ad-hoc one-liners and manual touch-ups are the diffuse mutation failure mode: no single point can be audited, re-run, or blamed. Finding: name every mutation path outside the chokepoint.
• Hash-witnessed backups before rewrite. Before the chokepoint runs, the originals are preserved with content hashes recorded (a committed baseline counts), so "the rewrite changed only what it claims" is checkable byte-for-byte, not asserted. Finding: a bulk rewrite with no verifiable before-state.
• Self-administered ambition gate. The change states what it deliberately does not touch, and the diff respects it. A formatter run that also renames, a codemod that also refactors, is the scope-creep rewrite failure mode. Finding: any file class in the diff outside the change's own stated scope.

Stop condition for this check: all three standards have an explicit pass or finding; a bulk-rewrite review that reports style nits but skips these is incomplete.

References

• Common standards
• Go
• Python
• Rust
• TypeScript
• JavaScript
• Shell
• JSON
• YAML
• Markdown
• SQL safety
• Race conditions
• LLM trust boundaries
• Skill structure
• Test strategy