CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

afk-retrospective

评估 AFK 批次的工程、产品、知识与执行改进,生成可留存报告供用户裁决。

Quality

74%

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

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/afk-retrospective/SKILL.md
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.

An exceptionally well-crafted instruction-only skill: lean prose, concrete file paths and verification steps, a clearly sequenced three-phase workflow with explicit stop conditions, disclosure requirements, a pre-delivery checklist, and explicit completion criteria. The only notable weakness is the absence of literal commands, which keeps actionability at 4 rather than 5.

DimensionReasoningScore

Conciseness

The body is dense, directive prose with zero explanation of concepts Claude already knows — no padding, no library intros, no generic filler. A few judgment-philosophy sentences (e.g. "候选是待验证的假设,不因被 reviewer 或上游 Agent 提出就值得做") could be tightened slightly, keeping it at anchor 4 rather than 5.

4 / 5

Actionability

Concrete guidance throughout: exact paths (`.afk/<batch-id>.jsonl`, `.afk/<batch-id>/handoff-*.md`, `.afk/reports/<batch-id>/<唯一文件名>.html`), specific checks (确认账本含 `closed` 行, 记录核验 commit SHA, 在可用浏览器中检查显示). Falls short of 5 only because no literal commands are given (e.g. `git fetch origin`) — acceptable for an instruction-only skill, but a minor gap.

4 / 5

Workflow Clarity

A clear three-phase sequence (建立事实 → 判断是否值得做 → 报告与裁决) with explicit validation checkpoints and error paths: stop if the ledger lacks a `closed` row, disclose when origin/main cannot be refreshed or claims cannot be verified, pre-delivery checks of body/figures/anchors/sources plus browser rendering with explicit handling of un-performable checks, and explicit completion criteria. The workflow is read-only ("仅允许新增本地报告"), so the destructive-cap does not apply.

5 / 5

Progressive Disclosure

The skill is under 50 lines, self-contained, and organized into clearly scoped sections. Its four cross-references ([lessons.md], [writing-for-agents], [domain-modeling], [escalation.md]) are one level deep and each clearly signaled with its purpose ("涉及 Agent 指令时参考..."). No bundle files exist to warrant further splitting.

5 / 5

Total

18

/

20

Passed

Description

53%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 clearly communicates what the skill does and occupies a distinct, project-specific niche, but it entirely lacks 'when to use' trigger guidance and natural trigger phrases. Although the skill is marked disable-model-invocation, the description still misses the rubric's explicit completeness and trigger-term requirements.

Suggestions

Append an explicit trigger clause, e.g. "Use when the user asks to retrospect/review a just-finished AFK batch (复盘/回顾/总结 an AFK 批次)."

Add natural synonyms users would actually say — 复盘, 回顾, 批次总结, handoff review — so trigger matching works on real phrasing, not just the 'AFK' jargon.

Briefly enumerate the concrete outputs (e.g., fact-verification against origin/main SHA, filtered numbered recommendations, offline single-file HTML report) to lift specificity beyond two generic verbs.

DimensionReasoningScore

Specificity

The description names its domain ("AFK 批次") and two concrete actions ("评估 AFK 批次的工程、产品、知识与执行改进", "生成可留存报告供用户裁决") but coverage is not comprehensive — verification, evidence-gathering, and report structure are absent. It sits at anchor 3 rather than 4 because it lists only two verbs rather than several specific actions.

3 / 5

Completeness

The 'what' is clear — evaluate engineering, product, knowledge, and execution improvements and generate a retained report for user adjudication — but there is no 'Use when...' clause or equivalent trigger guidance, capping completeness at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

Relevant keywords are present ("AFK 批次", "改进", "报告", "裁决") but common variations and synonyms a user might naturally say (e.g. 复盘/retrospective, 总结, handoff) are missing from the description. Not score 2 because multiple genuinely relevant, non-generic keywords exist rather than one or two generic ones.

3 / 5

Distinctiveness Conflict Risk

"AFK 批次" is a highly project-specific niche with minimal overlap risk against other skills. It falls short of anchor 5 only because distinct trigger phrases are not explicitly stated, leaving minor overlap risk with adjacent review/handoff-type skills.

4 / 5

Total

13

/

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-retrospective/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.