babysit-pr

Monitor a PR, fix feedback and CI failures until fully green for 30 min. Run with /babysit-pr <number>

Quality

64%

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/babysit-pr/SKILL.md

Quality

Content

63%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 content is highly actionable with a well-sequenced, validation-rich workflow, but it is monolithic and verbose with notable repetition of the same safeguards. Splitting the lease-coordination detail into a reference file and de-duplicating repeated rules would lift the two lowest dimensions.

Suggestions

Move the lease/heartbeat coordination logic into a separate reference file (e.g., LEASE.md) and summarize it one level deep in SKILL.md to improve progressive disclosure.

De-duplicate the repeated concurrent-work preservation and lease-renewal rules into a single canonical section referenced where needed.

Tighten the verbose Setup paragraph (e.g., consolidate the repeated "remain foreground-only" outcomes) to reduce token cost without losing the edge-case coverage.

Dimension	Reasoning	Score
Conciseness	The body is operationally dense rather than padded with concepts Claude already knows, but it repeats the same safeguards many times (concurrent-work preservation, lease renewal, "remain foreground-only") in a way that could be tightened, fitting anchor 3 rather than the lean anchor 4.	3 / 5
Actionability	It provides many concrete, copy-pasteable commands (git fetch/status, gh pr view/checks, the jq unaddressed-comments pipeline, the changeset format), but some guidance is conditional/abstract ("preserve concurrent work", "wait for its owner") rather than executable, so it stops short of fully copy-paste anchor 5.	4 / 5
Workflow Clarity	There is a clear sequenced workflow (Setup → Each tick Step 0/1 → PR checks → Merging → Stop conditions) with explicit validation checkpoints (status check before push, mergeable gate, 10-min soak, final unaddressed-comments gate) and feedback loops, but the lease-coordination sub-sequence is convoluted, leaving minor gaps short of anchor 5.	4 / 5
Progressive Disclosure	The skill is a single 385-line SKILL.md with no bundle files and no external references; sections are well-headed, but the large inline lease-coordination block is content that would be better split into a one-level-deep reference file, fitting anchor 3 rather than the well-split anchor 4.	3 / 5
	Total	14 / 20 Passed

Description

66%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 and distinct with good natural trigger terms, but lacks an explicit 'Use when...' clause so its completeness is capped. Adding a natural-language usage trigger would raise the weakest dimension.

Suggestions

Add an explicit 'Use when...' clause (e.g., 'Use when you need a PR kept green and fully reviewed for 30 minutes without manual babysitting').

Include synonyms like 'pull request' and 'CI checks' alongside 'PR' and 'CI failures' for broader trigger coverage.

Consider listing one or two more concrete actions (e.g., 'resolve merge conflicts', 'reply to review comments') to push specificity toward comprehensive.

Dimension	Reasoning	Score
Specificity	Names the PR domain and several concrete actions — "Monitor a PR, fix feedback and CI failures until fully green for 30 min" — with an explicit termination condition, which sits above anchor 3's 1–2 actions but short of the comprehensive coverage of anchor 5.	4 / 5
Completeness	It has a clear "what" (monitor PR, fix feedback and CI failures until green for 30 min) but no explicit "Use when..." trigger clause — the trigger is only the /babysit-pr invocation — so per the guideline a missing trigger clause caps completeness at 3.	3 / 5
Trigger Term Quality	Includes natural terms a user would say ("PR", "CI failures", "feedback", "fully green") plus the command trigger, but misses common synonyms like "pull request" or "checks", so it is not the comprehensive coverage of anchor 5.	4 / 5
Distinctiveness Conflict Risk	The 30-minute green-gate babysitting niche with a slash-command trigger is clearly distinct from related review/ship skills, with only minor overlap risk, fitting anchor 4 rather than the minimal-conflict anchor 5.	4 / 5
	Total	15 / 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
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: BuilderIO/agent-native
Path: .agents/skills/babysit-pr/SKILL.md
Commit: cc2a915

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.