CtrlK
BlogDocsLog inGet started
Tessl Logo

harness-release

Generic release automation for projects using Keep a Changelog + GitHub. Single confirmation gate then end-to-end automation: bump detection, CHANGELOG promotion, PR/main merge, tag, GitHub Release. Trigger: release, version bump, publish. Do NOT load for: implementation, review, planning, setup.

73

Quality

92%

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

The canonical home for this skill is harness-release in Chachamaru127/claude-code-harness

SKILL.md
Quality
Evals
Security

Quality

Content

85%

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

A well-engineered release skill: executable commands, clear single-gate sequencing with validation and failure recovery, and clean one-level-deep references. The main weakness is conciseness — several contextual/rationale passages could be trimmed or offloaded to references.

Suggestions

Tighten the 'CC runtime hard floor' and 'mirror drift / run ID' passages to one or two lines and move historical regression detail into a reference file, since it competes with context without changing the action.

Consolidate the repeated PR-ready/release-ready and bare-invocation rationale blocks that appear across the Quick Reference, bare invocation contract, and 設計思想 sections to remove near-duplicate explanation.

Consider moving the full Confirmation Gate ASCII plan block into post-gate-detail.md and summarizing the three-choice gate inline, so the overview body stays lean.

DimensionReasoningScore

Conciseness

The body is accurate and mostly purposeful but is long (~360 lines) and re-explains operational context Claude can derive (e.g. multiple paragraphs on runtime hard floor, mirror drift run IDs, lane taxonomy) that could be tightened or pushed to references.

2 / 3

Actionability

Provides concrete, copy-pasteable bash (preflight checks, `claude plugin tag --dry-run`, `check-release-version-sync.py --json`, verify-publish polling) and exact literal markers (RELEASE_AUTOSTART, the trailing output literal) rather than vague direction.

3 / 3

Workflow Clarity

The single-gate flow is explicitly sequenced as numbered Pre-Gate → Confirmation Gate → Post-Gate stages with a failure-recovery table, validation checkpoints (preflight fail-closed, version-sync re-check after merge), and a REQUEST_CHANGES→fix→re-review feedback loop.

3 / 3

Progressive Disclosure

Body is an overview that links one level deep to real reference files (version-files.md, bump-detection.md, release-notes.md, github-release.md, versioning.md, post-gate-detail.md), all verified present, with the detail split out rather than inlined.

3 / 3

Total

11

/

12

Passed

Description

100%

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 specific, trigger-rich, and complete: it states concrete release actions, natural trigger terms, and an explicit load/don't-load boundary. It is a strong, concise model description with no over-claims.

DimensionReasoningScore

Specificity

Lists multiple concrete actions in one sentence — "bump detection, CHANGELOG promotion, PR/main merge, tag, GitHub Release" — each naming a distinct release step.

3 / 3

Completeness

Clearly answers what ("end-to-end automation" across the named steps) and when with an explicit "Trigger: release, version bump, publish" clause plus a "Do NOT load for" exclusion list.

3 / 3

Trigger Term Quality

Includes natural user-facing terms "release, version bump, publish" that a user would actually say when needing this skill.

3 / 3

Distinctiveness Conflict Risk

Niche is clearly scoped to Keep-a-Changelog + GitHub release automation, and the explicit "Do NOT load for: implementation, review, planning, setup" anti-triggers sharply reduce conflict with sibling harness skills.

3 / 3

Total

12

/

12

Passed

Validation

87%

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

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 7 missing, 7 deeper-than-1-level

Warning

referenced_paths_exist

Referenced path issues: 6 missing

Warning

Total

14

/

16

Passed

Repository
Chachamaru127/claude-code-harness
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.