autoresearch-finalize

Finalize an autoresearch session into clean, reviewable branches. Use when asked to "finalize autoresearch", "clean up experiments", or "prepare autoresearch for review".

Quality

81%

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

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 tight, well-sequenced workflow with strong validation and feedback loops for a destructive batch operation. Its main weakness is that the pivotal finalize.sh script is referenced but absent from the bundle, leaving the core execution step unverifiable.

Suggestions

Bundle finalize.sh under scripts/ so the central execution step (bash <SKILL_DIR>/finalize.sh) is actually present and runnable, or inline its behavior as explicit commands.

De-duplicate the "no two groups may touch the same file" rule so it is stated once authoritatively rather than repeated across Step 1 and Step 2.

Consider collapsing the repeated "(legacy: ...)" filename annotations into a single legacy-mapping note to reduce clutter.

Dimension	Reasoning	Score
Conciseness	The body is largely efficient, assuming git competence and giving actionable rules rather than concept padding; minor redundancy (the file-disjoint rule repeats across Step 1 and Step 2) and the repeated "(legacy: ...)" annotations could be trimmed slightly.	4 / 5
Actionability	It provides concrete executable commands (git rev-parse, git merge-base, bash <SKILL_DIR>/finalize.sh) and a complete groups.json schema with a worked example, but the central execution step depends on finalize.sh, which is not present in any scripts/ bundle.	4 / 5
Workflow Clarity	A clear three-step sequence with explicit validation checkpoints ("Wait for approval before proceeding", script verifies the union tree, rollback on creation failure) and a feedback loop (overlapping files fail → merge groups and retry), plus an Edge Cases checklist.	5 / 5
Progressive Disclosure	Content is well-sectioned (Step 1/2/3, Edge Cases) and appropriately self-contained for its size, with the single external reference (finalize.sh) clearly signaled; the gap is that the referenced script is not actually bundled.	4 / 5
	Total	17 / 20 Passed

Description

82%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 strong: it covers a clear niche, explicitly states both capability and trigger conditions, and uses natural phrasing. The main weakness is specificity, which lists one main action rather than the several concrete operations the skill performs.

Suggestions

Expand the capability clause to list 2-3 concrete actions, e.g. "group kept experiments into independent changesets, expand short hashes, and create one branch per group from the merge-base".

Add a synonym or shorthand trigger (e.g. "split autoresearch branch" or "autoresearch branches") to broaden natural-term coverage.

Dimension	Reasoning	Score
Specificity	"Finalize an autoresearch session into clean, reviewable branches" names the domain and one concrete action (splitting into branches), but does not enumerate the several distinct sub-actions (grouping, hash expansion, merge-base branching, verification).	3 / 5
Completeness	It explicitly answers both what ("Finalize an autoresearch session into clean, reviewable branches") and when ("Use when asked to...") with concrete trigger phrases.	5 / 5
Trigger Term Quality	It provides three natural trigger phrases a user would actually say ("finalize autoresearch", "clean up experiments", "prepare autoresearch for review"), giving good coverage, though it lacks further synonyms or file-extension-style variants.	4 / 5
Distinctiveness Conflict Risk	"autoresearch" is a clear, narrow niche with distinct trigger phrases, giving it minimal overlap risk with other skills.	5 / 5
	Total	17 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: davebcn87/pi-autoresearch
Path: skills/autoresearch-finalize/SKILL.md
Commit: 4182079

Reviewed: 2 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.