CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

afk-team-workflow

把一个 Spec 的全部子 issue 或一批零散 issue 组建团队无人值守跑到全部合并或明确暂停。

Quality

82%

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.

An exceptionally well-engineered orchestration skill: lean rule-dense prose, copy-paste-ready commands and exact branch/merge procedures, and a clearly sequenced workflow with validation checkpoints, feedback loops, and stuck-detection for batch operations. The only structural weakness is navigation: the central handoff format doc is unlinked from the main file and several doc links point outside the bundle.

DimensionReasoningScore

Conciseness

The body is extremely dense with zero filler: no concept explanations, differences between batch types compressed into a table, and every sentence stating an operational rule (writer assignments, end conditions, label transitions), assuming Claude's competence throughout.

5 / 5

Actionability

Guidance is fully concrete: exact commands with flags ('scripts/batch-poll.sh --spec <N>', '--issues <N,...>', 'scripts/ledger.sh --new', '--scope-issues <N,...>'), branch naming 'afk/<batch-id>/stage-<K>', '--force-with-lease' push, expected-head rebase merge, and GitHub API fields ('headRefOid', 'mergeable=MERGEABLE'). Only trivial gap: scripts/repo-context.sh is never mentioned.

5 / 5

Workflow Clarity

The multi-stage process is explicitly sequenced (plan → serial stages with a writer/end-condition table for integration, review, PR → pause → wrap-up) with strong validation checkpoints (green HEAD, full-mode QA, a four-condition merge checklist) and genuine feedback loops (fixer → re-verify repetition until no failures or stuck detection '同一失败特征第 2 次出现'), so the batch-operation cap does not apply.

5 / 5

Progressive Disclosure

The overview links eight one-level-deep reference files (spawn-prompts, herdr-teammate, implementer, stage-reviewer, qa-verifier, review-looper, fixer, recovery), all present in the bundle, with content appropriately split. Minor gaps: references/handoff.md — a concept used throughout the body — is never linked from SKILL.md itself (only reachable via role docs), and four links (../../../docs/agents/*, ../model-routing/SKILL.md) point outside the bundle and do not resolve in it.

4 / 5

Total

19

/

20

Passed

Description

61%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 concise and names a precise, distinctive niche with the key natural terms (Spec, 子 issue, 无人值守, 合并, 暂停), but it covers only 1-2 concrete actions and lacks any explicit 'when to use' trigger guidance, capping completeness and specificity at 3. Note that disable-model-invocation: true makes trigger guidance less critical for this skill, but the rubric scores only what is written.

Suggestions

Add an explicit trigger clause, e.g. '用户要求无人值守/挂机把一个 Spec 的全部子 issue 或一批零散 issue 跑完时使用' — this would lift completeness from 3 toward 5 even though the skill is slash-invoked.

Enumerate one or two more concrete capabilities in the description (e.g. stage 划分与串行合并、PR 审查收敛与 QA 复验、部署冒烟) so specificity moves from '1-2 concrete actions' toward 'several specific actions'.

Include a natural synonym such as '挂机' or '自动跑' alongside '无人值守' to broaden trigger-term coverage.

DimensionReasoningScore

Specificity

Names the domain precisely ('一个 Spec 的全部子 issue 或一批零散 issue') and 1-2 concrete actions ('组建团队', '无人值守跑到全部合并或明确暂停'), but does not list the several specific actions the skill actually covers (stage planning, integration review, PR/QA loops, smoke testing), so the 'several specific actions' anchor is not met.

3 / 5

Completeness

The 'what' is clear (assemble a team and run a batch of issues unattended to full merge or explicit pause), but there is no explicit 'Use when...' clause or equivalent trigger guidance; 'when' is only weakly implied by the input specification, which caps completeness at 3 per the guidelines.

3 / 5

Trigger Term Quality

Contains the natural phrases a user of this workflow would say — 'Spec', '子 issue', '零散 issue', '无人值守', '合并', '暂停' — giving good keyword coverage, though common variations like '批次', '自动' or '挂机/AFK' are missing, keeping it below the comprehensive anchor.

4 / 5

Distinctiveness Conflict Risk

The composite of unattended team execution over Spec sub-issues or scattered issue batches to a merge/pause terminal state is a clear niche with low conflict risk, but neighboring agent-orchestration skills could still overlap since no distinct trigger phrases are stated.

4 / 5

Total

14

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 4 suspicious

Warning

Total

15

/

16

Passed

Repository

ArcReel/ArcReel

Path

.agents/skills/afk-team-workflow/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.