CtrlK
BlogDocsLog inGet started
Tessl Logo

fix-local-checks

Patch the code so `scripts/local-checks.sh` passes, after /implement-mainspec — a narrow post-implement polish specialist. Reads the (remediation-rich) check failures, fixes the underlying cause, never silences a check, re-verifies, and commits. Invoked headless by the dispatcher's two-strike local-checks gate; the dispatcher owns the retry counter. Triggers - fix-local-checks, fix lint, fix local checks, fix typecheck, make checks pass, post-implement polish (project)

75

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A tight, highly actionable policy skill: concrete commands, an explicit feedback loop, a well-specified escalation path, and exhaustive forbidden-action lists. Its only weakness is redundancy — the never-rules are restated across three sections (most notably the closing "Hard nevers" section), which costs tokens without adding guidance.

Suggestions

Merge the "Hard nevers" section into the earlier "You may NEVER silence a check" section and skip rule, which already state the same constraints — one authoritative list would cut ~15 lines with no loss of guidance.

Trim the motivational framing in "This is the whole reason you exist as a separate, constrained skill" and "A check passing is a proxy..." to a single rule statement; the constraint is clear without the justification prose.

Consider moving the dispatcher contract details (invocation mechanics, counter file path, tick/STUCK lifecycle) into a short reference file, keeping the SKILL.md body focused on the fix workflow itself.

DimensionReasoningScore

Conciseness

The body is efficient and assumes Claude's competence (no explanation of what linting or typechecking is), but the "never" rules are stated three times: the "You may NEVER silence a check" section, the skip rule's restatement, and the closing "Hard nevers" section which nearly duplicates prior content ("Never silence a check... Never add a test-skip marker... Never touch the check machinery"). Not 5 because the Hard nevers section is redundancy that could be trimmed; not 3 because the padding is minor and deliberate (load-bearing constraints) rather than unnecessary explanation.

4 / 5

Actionability

Fully concrete guidance: exact commands (`./scripts/local-checks.sh`, `claude -p "/fix-local-checks <feature>"`), a specific commit message example (`fix(local-checks): trim query whitespace`), an exhaustive catalog of forbidden suppression directives (eslint-disable, @ts-ignore, # noqa, #[allow(...)]), and the exact counter file path (.harness/local-check-attempts-<f>). Not 4 because the common cases are covered copy-paste ready with no gaps.

5 / 5

Workflow Clarity

Clear 5-step sequence (read failures → triage → fix cause → re-verify → commit/push) with an explicit validation checkpoint ("Re-run `./scripts/local-checks.sh`" before commit), a feedback loop (commit only honestly green or honest partial progress so attempts compound), and a well-defined error-recovery path (step 5: stop, commit partial progress, let the dispatcher STUCK). Not 4 because validation and recovery are explicit, not implicit.

5 / 5

Progressive Disclosure

No bundle files exist, so the skill is appropriately self-contained: nothing (API reference, catalogs) is bulk-inlined that clearly belongs in a separate file, and sections are well-organized with clear headers. Not 5 because at ~105 lines the supporting detail (dispatcher contract mechanics, the suppression-directive catalog) could be externalized to references to keep the core workflow lean; not 3 because structure and navigation are good with no buried content.

4 / 5

Total

18

/

20

Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is exemplary: it names the exact script and workflow, lists concrete actions, draws an explicit boundary against the sibling implement skill, and provides an explicit trigger list with natural synonyms. Voice is third person and there is no fluff or over-claiming.

DimensionReasoningScore

Specificity

Multiple concrete, specific actions are listed: "Patch the code so `scripts/local-checks.sh` passes", "Reads the (remediation-rich) check failures, fixes the underlying cause, never silences a check, re-verifies, and commits" — comprehensive coverage with a named target script. Not level 4 because coverage is complete rather than having minor gaps; there is no vague or padded language.

5 / 5

Completeness

Both "what" (patch code so local-checks passes by fixing underlying causes, never silencing checks, re-verify, commit) and "when" ("after /implement-mainspec", "Invoked headless by the dispatcher's two-strike local-checks gate", plus an explicit "Triggers - ..." clause) are clearly and explicitly stated. Not level 4 because the "when" is fully explicit with concrete trigger phrases rather than merely present.

5 / 5

Trigger Term Quality

The explicit trigger list covers natural phrasings and their variations: "fix-local-checks, fix lint, fix local checks, fix typecheck, make checks pass, post-implement polish" — these are exactly what a user/dispatcher would say when this skill is needed. Not level 4 because synonym coverage (lint/typecheck/checks) is thorough with no obvious natural term missing.

5 / 5

Distinctiveness Conflict Risk

A clear niche with distinct triggers: "a narrow post-implement polish specialist" scoped to `scripts/local-checks.sh` after /implement-mainspec, explicitly contrasted with "not a second `/implement-mainspec`". Minimal conflict risk — the trigger phrases map to this skill alone. Not level 4 because the boundary against the neighboring implement skill is drawn explicitly, leaving only minimal overlap.

5 / 5

Total

20

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 4 missing

Warning

Total

15

/

16

Passed

Repository
tdg-ninja/context-specs-factory-ai
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.