pr-gardening

Discover the pull requests this Paperclip instance opened (never community contributions), report what each is for and how confident we are that it is merge-ready, and automatically drive the non-ready ones back to green with /prepare-paperclip-pr — without ever merging.

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/pr-gardening/SKILL.md

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 strong operational skill: every stage is driven by concrete, executable commands with exact API endpoints and templates, and the workflow is clearly sequenced with genuine validation checkpoints and feedback loops. The only real weaknesses are a monolithic inline body with no reference files for detailed material and some repetition of the safety guardrails.

Dimension	Reasoning	Score
Conciseness	The body is dense and imperative with no explanations of concepts Claude already knows, but safety rules like 'never merge, approve, or close' are restated across the Hard Guardrails, Stage C, and Stage F sections — minor repetition that could be trimmed, placing it at 'efficient; minor instances of over-explanation' rather than 'every token earns its place'.	4 / 5
Actionability	Fully executable throughout: runnable `node .agents/skills/pr-gardening/scripts/find-candidates.mjs`-style commands with flags, exact API endpoints (`GET /api/companies/:companyId/search/extract`, `POST /api/issues/:issueId/comments`), idempotency-key formats, and a copy-paste-ready marker-comment template covering the common cases.	5 / 5
Workflow Clarity	Stages A through F are clearly sequenced with explicit validation checkpoints and feedback loops: 'Always rerun this stage after any wake', 'Rerun Stage B immediately before this stage', and a Verification section with `node --test` plus a live dry-run sanity checklist (community PRs must appear only under `droppedCommunityPullRequests`). The destructive/batch cap does not apply since validation is present.	5 / 5
Progressive Disclosure	Script logic is appropriately externalized to real bundle files that are all correctly referenced inline (find-candidates.mjs, check-readiness.mjs, render-report.mjs, pr-gardening.test.mjs all exist in ./scripts/), and sections are well organized. However, the body is a ~180-line monolith with no one-level-deep reference docs for detailed material like the deduplication rules and API contracts, matching 'good structure; minor organization gaps' rather than a reference-pointing overview.	4 / 5
	Total	18 / 20 Passed

Description

62%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 concrete, domain-specific, and clearly delineated from other skills, but it lacks any 'Use when...' trigger clause, misses natural synonyms like 'PR' and 'gardening', and slips into first person ('how confident we are'). It communicates the what well while leaving the when largely implicit.

Suggestions

Add an explicit trigger clause, e.g. 'Use when asked to garden, triage, or check the readiness of this instance's PRs or when maintaining pull requests opened by Paperclip.'

Include natural synonyms users would say — 'PRs', 'pull request gardening', 'CI green' — to improve trigger term coverage.

Rewrite in consistent third person ('...and reports how confident each is that it is merge-ready') to remove the first-person 'how confident we are'.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('Discover the pull requests this Paperclip instance opened', 'report what each is for and how confident we are that it is merge-ready', 'drive the non-ready ones back to green with /prepare-paperclip-pr') with comprehensive domain coverage, matching the 5 anchor — but the first-person phrase 'how confident we are' violates the third-person-voice guideline, reducing the score by 1.	4 / 5
Completeness	The 'what' is clear and specific (discover, report readiness confidence, drive to green without merging), but there is no 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	Relevant keywords like 'pull requests', 'merge-ready', and 'drive back to green' are present, but common variations users would naturally say — 'PR', 'gardening', 'CI' — are missing, matching the 'some relevant keywords but missing common variations or synonyms' anchor rather than the good-coverage anchor above.	3 / 5
Distinctiveness Conflict Risk	The description is anchored to a clear niche — 'the pull requests this Paperclip instance opened (never community contributions)' and '/prepare-paperclip-pr' — giving it distinct, tool-specific triggers with minimal overlap risk against other skills.	5 / 5
	Total	15 / 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
relative_links	Relative link issues: 1 suspicious	Warning

	Total	15 / 16 Passed

Repository: paperclipai/paperclip
Path: .agents/skills/pr-gardening/SKILL.md
Commit: 1c07b59

Reviewed: 1 day 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.