CtrlK
BlogDocsLog inGet started
Tessl Logo

batch-bug-shepherd

Use this skill to drive a batch of suspected bugs in microsoft/apm from raw issue list to mergeable PR queue. Fan out one triage subagent per issue (LEGIT / UNCLEAR / FIXED-AT-HEAD), gate every legit bug against PRINCIPLES.md via an apm-ceo strategic-alignment pass, cross-reference legit issues against open PRs, then open a fix PR (TDD + mutation-break gate) for greenfield bugs. Drive every PR -- community in-flight and own fix alike -- to mergeable by composing the shepherd-driver skill: one driver per PR runs the review panel, folds non-blocking recommendations, pushes (preserving author), and watches CI to green. Re-probe mergeability and resolve conflicts via shepherd-driver. Maintain a plan.md ground-truth table as canonical state. Activate when the maintainer asks to triage issues, sweep the bug queue, shepherd bug-flagged issues, run a weekly community sweep, or drive in-flight community PRs to merge -- even if "shepherd" or "batch" is not named.

73

Quality

92%

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

85%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, highly actionable orchestrator body with excellent workflow sequencing, validation gates, and progressive disclosure that loads detail one level deep. Its one weakness is conciseness: binding contracts and the 18 invariants are restated across multiple sections and duplicated in the reference file, adding tokens that could be trimmed.

Suggestions

Trim the inline 18-invariant list to the few that change behavior per phase and rely on references/invariants.md for the rest, keeping only the 'load before Phase 0' pointer — the full set is already duplicated there.

State each cross-cutting contract (worktree isolation, lint-as-push-gate, probe-for-shepherd-driver) once at first use and reference it later instead of restating it in the Composition, invariant, phase, and operating-contract sections.

Replace ALL-CAPS emphasis ('WILL', 'NEVER', 'MUST', 'NOT') with short imperative sentences; the caps add visual weight without adding information Claude needs.

DimensionReasoningScore

Conciseness

The body is information-dense with no beginner-concept padding, but the 18 invariants are summarized inline AND fully duplicated in references/invariants.md, and contracts (worktree isolation, lint-as-push-gate, probe-for-shepherd-driver) are restated across the Composition, invariant, phase, and operating-contract sections, so there is more than minor tightening opportunity — fitting score 3 over 4.

3 / 5

Actionability

Provides copy-paste-ready, executable commands — 'gh issue list --label bug --state open --json ...', 'git worktree add <path> origin/main', the 'uv run python <row-worktree>/.agents/skills/shepherd-driver/scripts/owner_touch_gate.py verify --base $BASE_SHA --head ...' block, and the test-ff probe snippet — plus concrete schemas and status enums, matching the fully-executable score-5 anchor.

5 / 5

Workflow Clarity

Phases 0-6 are clearly sequenced with explicit validation checkpoints (schema-validate every return with retry-once, mutation-break gate, owner_touch_gate parent re-probe, mergeability re-probe, lint-as-push-gate) and error-recovery feedback loops; the batch/destructive cap does not apply because validation is robust, matching the score-5 anchor.

5 / 5

Progressive Disclosure

The body is an overview that points one level deep to real, well-signaled references — 'Load references/invariants.md before planning Phase 0', 'Load references/strategic-alignment-gate.md when entering this phase', '../shepherd-driver/references/mergeability-gate.md' — and appropriately delegates PR-drive detail to the sibling skill rather than duplicating it; all referenced bundle files exist, matching the score-5 anchor.

5 / 5

Total

18

/

20

Passed

Description

100%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.

An exemplary description: it names a precise niche, enumerates concrete actions across every wave, and gives explicit multi-phrase activation triggers with a synonym catch. It cleanly separates itself from the composed sibling skills.

DimensionReasoningScore

Specificity

Lists many concrete actions — 'fan out one triage subagent per issue (LEGIT / UNCLEAR / FIXED-AT-HEAD)', 'gate every legit bug against PRINCIPLES.md', 'open a fix PR (TDD + mutation-break gate)', 'folds non-blocking recommendations, pushes (preserving author), and watches CI to green' — with comprehensive coverage, matching the score-5 anchor rather than the partial list at 4.

5 / 5

Completeness

Explicitly states what ('drive a batch of suspected bugs ... from raw issue list to mergeable PR queue' through the full wave chain) and when ('Activate when the maintainer asks to ...'), with concrete trigger phrases, hitting the score-5 anchor for both what and when.

5 / 5

Trigger Term Quality

Gives five natural maintainer phrases — 'triage issues, sweep the bug queue, shepherd bug-flagged issues, run a weekly community sweep, or drive in-flight community PRs to merge' — plus the 'even if shepherd or batch is not named' synonym catch, matching the comprehensive-coverage anchor at 5.

5 / 5

Distinctiveness Conflict Risk

Occupies a clear niche — batch bug-queue orchestration in microsoft/apm that composes shepherd-driver — with distinct triggers and an 'Out of scope' boundary against single-PR shepherd-driver and apm-review-panel, so conflict risk is minimal, matching the score-5 anchor.

5 / 5

Total

20

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 2 suspicious

Warning

referenced_paths_exist

Referenced path issues: 2 missing

Warning

Total

14

/

16

Passed

Repository
microsoft/apm
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.