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 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is a strong, highly actionable orchestrator spec with concrete commands, explicit validation checkpoints, and well-signaled one-level-deep references whose files all exist. Its only weakness is conciseness: the driver return enum, probe block, and composition rationale are restated across multiple sections and could be consolidated.

Suggestions

Consolidate the repeated shepherd-driver composition detail: state the completion_return enum and probe block once in 'Composition with shepherd-driver' and have Phases 4/5/Bundled assets reference it instead of restating.

The Architecture invariants section already defers full text to references/invariants.md; apply the same dispatch-anchor treatment to the per-phase driver/spawn contracts to cut body length.

Move the duplicated owner_touch_gate.py invocation pattern into a single canonical block and reference it from Phase 4 rather than re-deriving the command inline.

DimensionReasoningScore

Conciseness

The body is operational and free of concept-explainer padding, but restates the same information across sections (the completion_return enum, the shepherd-driver probe block, and the composition description recur in Composition, Phase 4, Phase 5, and Bundled assets), so it could be tightened rather than being fully lean.

2 / 3

Actionability

Provides fully executable, copy-paste-ready commands (gh issue/pr list with exact --json fields, git worktree add, the owner_touch_gate.py verify invocation) and explicit field derivations like HEAD_REPO = headRepositoryOwner.login + '/' + headRepository.name.

3 / 3

Workflow Clarity

Phases 0-6 are explicitly ordered with WAVE labels, dispatch-table-before-fan-out and re-render checkpoints, schema-validation with retry-once, and trust-but-verify re-probes; validation is pervasive for this batch operation so no cap applies.

3 / 3

Progressive Disclosure

Overview body uses one-level-deep references, each clearly signaled with phase-triggered load instructions (e.g. 'Load references/invariants.md before planning Phase 0'), and all 9 referenced assets/references files exist; the skill deliberately delegates PR-drive detail to the sibling rather than duplicating it.

3 / 3

Total

11

/

12

Passed

Description

100%Weight 40%Scale 1-3

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 exemplary: it states a comprehensive concrete action chain, gives explicit natural-language activation triggers, and carves out a distinct niche with named composed skills. It answers both 'what' and 'when' without verbosity or first/second-person voice issues.

DimensionReasoningScore

Specificity

Lists multiple concrete actions (triage fan-out with LEGIT/UNCLEAR/FIXED-AT-HEAD verdicts, PRINCIPLES.md strategic-alignment gate, PR cross-reference, TDD+mutation-break fix PR, drive-to-merge, conflict resolution, plan.md table), matching the 'lists multiple specific concrete actions' anchor.

3 / 3

Completeness

Clearly answers both 'what' (the full triage-to-merge action chain) and 'when' via an explicit 'Activate when...' trigger clause, satisfying the top anchor rather than leaving 'when' implied.

3 / 3

Trigger Term Quality

The explicit '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' clause covers natural maintainer phrasings and even handles the missing-keyword case.

3 / 3

Distinctiveness Conflict Risk

Niche is sharply defined around apm bug-queue orchestration with named composed skills (apm-ceo, shepherd-driver, mutation-break gate, ground-truth table), making it unlikely to trigger for the wrong skill.

3 / 3

Total

12

/

12

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.