story-done

End-of-story completion review — verifies each acceptance criterion, checks GDD/ADR deviations, prompts code review, updates status.

Quality

77%

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 ./.claude/skills/story-done/SKILL.md

Quality

Content

77%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.

An exceptionally actionable, well-sequenced workflow skill: exact commands, prompts, templates, verdict precedence, and validation gates throughout. Its weaknesses are length — ~780 lines with the qa.level/testing.strict policy explained twice and rationale commentary that could be trimmed (including one editing artifact at line 129) — and a monolithic single-file structure that inlines content a reference file should carry.

Suggestions

State the qa.level/testing.strict resolution rules once (Phase 1 or Phase 3) and cross-reference from the other phase instead of re-explaining them at length.

Move stable, rarely-changed material — the evidence-gate table, verdict definitions, and the tech-debt register row format — into a reference file (e.g. references/evidence-gates.md) to cut the body's length and duplication.

Fix the editing artifact at line 128–129 where "that should be in localization files" dangles after the bolded code-root clause, restoring the original 'Grep for player-facing strings that should be in localization files' instruction.

Dimension	Reasoning	Score
Conciseness	The body is dense and project-specific rather than padded with concepts Claude already knows, but the qa.level/testing.strict policy is expounded at length twice (Phase 1 and Phase 3), run-result rules repeat across phases, and design-rationale commentary plus an editing artifact (the dangling "that should be in localization files" at line 129) leave more than minor trimming. Anchor 3 rather than 4 because the tightening opportunities are substantial, but well above the verbosity of anchor 2.	3 / 5
Actionability	Guidance is fully concrete: exact Grep patterns and output modes, exact AskUserQuestion prompts and option labels, exact report/table/checkpoint templates, exact commands (e.g. `bash .claude/scripts/story-status.sh`) and exact tech-debt row formats. As an instruction-only skill this is copy-paste-ready guidance covering the common cases, matching anchor 5.	5 / 5
Workflow Clarity	Eight explicitly sequenced phases with validation before every write (report presented before any file edit, explicit user approval in Phase 7), feedback loops on failure (BLOCKED does not auto-proceed, "supply it and re-run the gate"), and explicit verdict precedence with definitions — a textbook match for anchor 5. The corrupted hardcoded-strings bullet muddies one check but does not degrade the sequence or validation structure.	5 / 5
Progressive Disclosure	Sections are well-organized and external doc references are clearly signaled one level deep, but the skill is a single ~780-line file with no bundle at all — evidence-gate tables, the config-resolution policy exposition, and the tech-debt row format duplicated from another skill are inlined where separate reference files belong. Anchor 3 rather than 4 because the volume of inlinable content is substantial; not 2 because structure and reference signaling are genuinely good.	3 / 5
	Total	16 / 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.

A terse, highly specific description that names four concrete actions and carves out a distinct niche in the story workflow. Its one real weakness is the missing explicit 'when to use' clause — timing is only implied by 'End-of-story' — which caps completeness at 3.

Suggestions

Add an explicit trigger clause, e.g. "Use when a story's implementation is finished and it needs to be closed, or when the user says a story is done / ready to mark complete."

Include natural synonyms users would say — "close the story", "mark the story complete", "finish a story" — to strengthen trigger-term coverage.

Clarify that it prompts (rather than performs) code review to sharpen the boundary with a standalone code-review skill.

Dimension	Reasoning	Score
Specificity	Lists four distinct concrete actions ("verifies each acceptance criterion, checks GDD/ADR deviations, prompts code review, updates status") covering the full completion-review workflow, matching the comprehensive-coverage anchor rather than anchor 4's 'minor gaps'.	5 / 5
Completeness	The 'what' is clear and multi-part, but there is no 'Use when...' clause — timing is only weakly implied by "End-of-story", which is exactly the anchor 3 case and is capped there by the judging guideline. Not anchor 2 because the 'what' is concrete, and not anchor 4 because the 'when' is never explicitly stated.	3 / 5
Trigger Term Quality	Good keyword coverage (story, completion, acceptance criteria, deviations, code review, status) but missing natural variations users would say such as "mark complete", "close the story", or "finish". Coverage is solid enough to sit above anchor 3 but lacks the synonym breadth of anchor 5.	4 / 5
Distinctiveness Conflict Risk	The story-completion/GDD/ADR traceability niche is mostly distinct, but "prompts code review" and the generic word "review" carry minor overlap risk with a code-review skill and neighboring story-workflow skills — anchor 4 rather than 5, and clearly more distinct than anchor 3's broad overlap.	4 / 5
	Total	16 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (794 lines); consider splitting into references/ and linking	Warning
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: Donchitos/Claude-Code-Game-Studios
Path: .claude/skills/story-done/SKILL.md
Commit: b21fa0f

Reviewed: 3 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.