babysit

Drive a PR to a clean review (Greptile 5/5, zero open threads) — ships if needed, keeps it mergeable against staging, triggers Greptile/Cursor Bugbot, fixes real findings, replies to and resolves every thread, and loops until clean

Quality

91%

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

Quality

Content

92%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 executable, well-structured review-loop workflow with explicit validation feedback loops and concrete commands throughout. The only weak spot is a few inline gotcha asides written as prose paragraphs that could be tightened for token efficiency.

Suggestions

Tighten the inline command-annotation asides (e.g. the `.body | tail -1` warning at the step-1 block and the `--reverse` note at step 7) from multi-sentence prose to one-line parentheticals so they earn their tokens without padding.

Consider moving the longer jq/git gotcha explanations into a short 'Pitfalls' subsection or a reference file so the core loop steps read as terse, scannable commands.

Dimension	Reasoning	Score
Conciseness	Lean and assumes Claude's competence (no explaining what a PR/Greptile/thread is), but several inline gotchas are written as multi-sentence prose paragraphs (e.g. the `.body \| tail -1` warning and the `--reverse` justification) that could be tightened to one line each.	4 / 5
Actionability	Fully executable throughout: concrete `gh`/`git`/GraphQL commands with real flags, exact API queries, copy-pasteable re-trigger mentions, and a specific ScheduleWakeup fallback delay, covering the common cases (merge conflict, fresh PR, open threads, false positive, already-fixed).	5 / 5
Workflow Clarity	Clear multi-step sequence with explicit validation checkpoints and feedback loops: check-state-before-acting, conflict-first ordering, sync check before every push, post-push verify after every push, re-trigger/wait/loop, and explicit stop conditions gated by the definition-of-clean contract.	5 / 5
Progressive Disclosure	Self-contained single file with clear `##` sections (When to use, Inputs, Definition of clean, Loop, Reporting, Hard rules), easy navigation, and one-level-deep references to the sibling `/ship` skill that are clearly signaled rather than nested.	5 / 5
	Total	19 / 20 Passed

Description

90%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 specific, tool-anchored description that clearly conveys both purpose and trigger state with comprehensive concrete actions. Minor room to surface a few more natural user synonyms as trigger terms.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('ships if needed', 'triggers Greptile/Cursor Bugbot', 'fixes real findings', 'replies to and resolves every thread', 'loops until clean') with a precise success criterion ('Greptile 5/5, zero open threads'), giving comprehensive coverage of what the skill does.	5 / 5
Completeness	Clearly answers 'what' (drive a PR to clean review via the listed actions) and 'when' is conveyed by 'Drive a PR to a clean review' as an explicit purpose with a concrete trigger state, matching the anchor that answers both with concrete trigger phrases.	5 / 5
Trigger Term Quality	Strong natural phrasing users would say ('babysit this PR', 'keep working the reviews until it's clean' appear in the body and 'Drive a PR to a clean review' is the trigger), plus named tools (Greptile, Cursor Bugbot), but the description itself lacks common synonyms like 'review comments' or 'mergeable' as natural user phrases.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (automated PR review-loop babysitting to Greptile 5/5 + zero open threads) tied to named tools (Greptile, Cursor Bugbot), with distinct triggers unlikely to fire for unrelated skills.	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: simstudioai/sim
Path: .agents/skills/babysit/SKILL.md
Commit: 804a452

Reviewed: 3 minutes 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.