CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

pr-shepherd

Takes a finished eve branch from review to merge. Opens the PR early, runs a fresh-context reviewer subagent, fixes critical findings, addresses, replies to, and resolves review comments, pushes, monitors CI (rerunning flaky jobs or fixing real failures), then squash-merges or enables auto-merge. Use after the initial implementation is complete, or when asked to open, babysit, land, or "get a PR green".

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

96%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.

Excellent operational content: copy-paste-ready gh/GraphQL commands, a clear seven-step workflow with validation checkpoints, failure-classification rules, and feedback loops, plus non-obvious repo-specific conventions throughout. The only structural improvement would be moving some heavier reference material (e.g., the GraphQL queries) into a references file to keep SKILL.md as a leaner overview.

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence: no generic git/gh/CI tutorials, and every section carries non-obvious repo-specific knowledge (bot review cadences, flake sources, DCO/signing rules, reply conventions, 'main does not require branches to be up to date'). Matches the 'every token earns its place' anchor.

5 / 5

Actionability

Fully executable throughout: complete GraphQL queries with --jq filters, concrete merge commands with flags ('gh pr merge <n> --squash --match-head-commit <headRefOid> --body "Signed-off-by: ..."'), 'gh run rerun <run-id> --failed', and copy-paste reply templates for the common cases.

5 / 5

Workflow Clarity

Steps 1-7 are clearly sequenced with explicit validation checkpoints (rerun fast checks, verify findings before acting, classify CI failures before rerunning, step 6's stop checklist, match-head-commit pin, post-merge confirmation) and genuine feedback loops (real CI failure -> fix -> back to step 4; 'Loop until done').

5 / 5

Progressive Disclosure

Well-organized sections and clearly signaled one-level-deep links (AGENTS.md, gh-pr-description/SKILL.md), but this ~250-line skill has no bundle files and keeps all operational detail inline (e.g., the GraphQL thread-harvest queries and CI failure triage), which fits the 'minor organization gaps' anchor rather than 'content appropriately split'.

4 / 5

Total

19

/

20

Passed

Description

92%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 strong description: third-person voice, comprehensive concrete actions covering the whole review-to-merge lifecycle, and an explicit 'Use when' clause with natural trigger phrasings. The only gap is that a few common synonyms users might say (merge, ship, close out) are not listed as triggers.

DimensionReasoningScore

Specificity

The description enumerates the entire PR-landing lifecycle as concrete actions ('Opens the PR early, runs a fresh-context reviewer subagent, fixes critical findings, addresses, replies to, and resolves review comments, pushes, monitors CI... then squash-merges or enables auto-merge'), matching the comprehensive-coverage anchor; the 4 anchor's 'minor gaps' does not apply since no lifecycle stage is missing.

5 / 5

Completeness

Both questions are answered explicitly: a detailed 'what' (the full action list) and a concrete 'when' clause ('Use after the initial implementation is complete, or when asked to open, babysit, land, or "get a PR green"'), matching the anchor that requires both with concrete trigger phrases.

5 / 5

Trigger Term Quality

Natural trigger phrases are present ('when asked to open, babysit, land, or "get a PR green"'), but common user phrasings such as 'merge my PR', 'ship it', or 'close out the PR' are not covered, so it fits the good-coverage-with-a-few-missing-terms anchor rather than the comprehensive-synonyms anchor.

4 / 5

Distinctiveness Conflict Risk

It occupies a clear niche (post-implementation PR landing for the eve repo, with repo-specific details like 'squash-merges or enables auto-merge' and dedicated triggers), so it is unlikely to fire for unrelated skills.

5 / 5

Total

19

/

20

Passed

Validation

81%

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

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

relative_links

Relative link issues: 2 suspicious

Warning

Total

13

/

16

Passed

Repository

vercel/eve

Path

.agents/skills/pr-shepherd/SKILL.md

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.