finish-feature

Finish a feature — run verification, an adversarial review gate (re-grounded on the original design doc + ticket ACs, not the plan) and a deferral-closure gate, then create the PR, retain learnings, and hand off cleanup to the terminal.

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/finish-feature/SKILL.md

Quality

Content

81%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.

The body is a high-quality, actionable, multi-step pipeline with strong validation checkpoints and feedback loops — a model for destructive/batch workflow clarity. Its only notable weakness is conciseness, driven by repeated restatement of the same guardrail warnings and lengthy rationale prose.

Suggestions

De-duplicate the 'align against the original design doc + ticket ACs, never the plan' warning — state it once authoritatively and reference it from the other locations rather than re-explaining each time.

Trim or collapse the long 'Why this step exists' blockquote in step 4 into a one-line rationale, keeping the detailed reasoning only where it changes behavior.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient operational guidance, but it restates the 'never align to the plan / the plan is a lossy compression' warning at least three times (step 4a, step 4 intro, and the Important section) and carries long 'Why this step exists' rationale blockquotes that could be tightened, matching the 3 anchor of mostly-efficient-with-some-unnecessary-explanation.	3 / 5
Actionability	Provides concrete, copy-paste-ready bash commands, a pinned commit-message regex, exact PR section headings, and a deterministic count block — fully executable guidance covering the common cases, matching the 5 anchor.	5 / 5
Workflow Clarity	A 9-step sequenced pipeline with explicit validation checkpoints and feedback loops — branch guard hard stop, 'If checks fail, report and stop', the pinned findings table gate, fail-loud on an empty diff range, and user-acceptance gates — exemplary for destructive/batch operations, matching the 5 anchor.	5 / 5
Progressive Disclosure	Well-organized into clear sections (Preamble, Determinism, Steps 1–9, Important) with clearly signaled external references (~/.claude/lib/determinism/CONTRACT.md and functions/*.sh); it is over 50 lines and inline rather than split into bundle files (references/scripts/assets are empty), so it sits at good structure with minor organization gaps rather than the ideal 5.	4 / 5
	Total	17 / 20 Passed

Description

71%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 specific and action-rich, clearly conveying a multi-phase feature-finishing pipeline. Its main weakness is the absence of an explicit 'Use when...' trigger clause, which caps completeness despite strong specificity and distinctiveness.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when a feature branch is implementation-complete and ready for verification, review, and PR creation.'

Include common trigger synonyms/variations (e.g. 'finalize feature', 'open a PR', 'wrap up a feature branch') to broaden natural keyword coverage.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'run verification, an adversarial review gate ... a deferral-closure gate, then create the PR, retain learnings, and hand off cleanup' — with comprehensive coverage of the pipeline, matching the 5 anchor.	5 / 5
Completeness	The 'what' is clear and detailed, but there is no explicit 'Use when...' clause or equivalent trigger guidance — it is only weakly implied by 'Finish a feature' — so per the judging guideline a missing explicit trigger caps completeness at 3.	3 / 5
Trigger Term Quality	'Finish a feature' is a natural phrase a user would say, with supporting terms (PR, review gate, verification), but it lacks synonyms or common variations of the trigger, so it sits above the 3 anchor but short of comprehensive 5 coverage.	4 / 5
Distinctiveness Conflict Risk	It carves a clear dev-flow 'finish a feature' niche with distinct triggers, but references domain-internal concepts (design doc + ticket ACs, deferral-closure gate) that may overlap with sibling pipeline skills, placing it mostly-distinct with minor overlap risk rather than minimal.	4 / 5
	Total	16 / 20 Passed

Validation

100%

Warnings & errors only

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: AndreJorgeLopes/devflow
Path: skills/finish-feature/SKILL.md
Commit: bd4b70c

Reviewed: 17 days ago

Table of Contents

Discovery Implementation Validation

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.