CtrlK
BlogDocsLog inGet started
Tessl Logo

lark-vc-agent

仅当用户或上游配置显式指定 lark-vc-agent 时使用,相关请求统一交由 lark-meeting 技能处理。

46

Quality

51%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/lark-vc-agent/SKILL.md

The canonical home for this skill is lark-minutes in larksuite/cli

SKILL.md
Quality
Evals
Security

Quality

Content

72%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.

As a minimal compatibility shim, the body is appropriately lean and correctly routes all real work to a clearly referenced sibling skill, with no padding of concepts Claude already knows. Its main weakness is the absence of any verification checkpoint confirming the referenced sibling skill exists and was loaded.

Suggestions

Add a brief verification step after the MUST-read directive, e.g. 'If ../lark-meeting/SKILL.md cannot be found, stop and inform the user the lark-meeting skill is missing.'

Clarify that no local processing should occur even if the request seems self-contained, to prevent drift off the router behavior.

Consider folding the redundant capability line ('本技能只用于兼容旧名称...') into the description so the body can be directive-only.

DimensionReasoningScore

Conciseness

The body is extremely short and free of concept explanations Claude already knows, with every line earning its place; it is not a 5 only because '本技能只用于兼容旧名称,不直接处理业务' slightly restates what the description already conveys.

4 / 5

Actionability

It gives a concrete, executable instruction ('MUST 完整读取 ../lark-meeting/SKILL.md 并按照其中的路由和行动指南执行') with a real file path; for this instruction-only router skill absence of code is not penalized, but it is not a 5 because there is no copy-paste-ready code or command.

4 / 5

Workflow Clarity

For a single-action router skill the single action (read the sibling SKILL.md and follow it) is unambiguous; the simple-skill exception applies and no destructive or batch operation is present, so the cap of 3 does not apply. It is a 4 rather than 5 because there is no explicit checkpoint confirming the sibling skill was actually located and read before proceeding.

4 / 5

Progressive Disclosure

The body stays a lean overview and offloads all detail to the one-level-deep reference ../lark-meeting/SKILL.md with a clearly signaled link; no references/scripts/assets bundle directories exist in this skill, and it is a 4 rather than 5 because the single reference points to a sibling skill rather than a well-organized bundle of topic files, limiting navigation breadth.

4 / 5

Total

16

/

20

Passed

Description

30%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 the skill's narrow compatibility-routing role and when to invoke it, which keeps conflict risk low. However, it is sparse on concrete actions and natural trigger terms, leaving both capability specificity and completeness below what the rubric rewards.

Suggestions

Add an explicit 'Use when ...' clause with concrete trigger phrases a user would naturally say, e.g. 'Use when the user explicitly asks for lark-vc-agent or references the legacy VC agent name'.

Name concrete actions rather than only 'route to lark-meeting', e.g. 'Forwards meeting/VC requests to the lark-meeting skill and otherwise performs no direct processing'.

Include natural synonyms for the legacy name to improve trigger-term coverage (e.g. 'lark-vc-agent', 'VC agent', 'lark video conference agent').

DimensionReasoningScore

Specificity

The description names the domain (lark-vc-agent compatibility) and one concrete action (delegating to lark-meeting), but actions are minimal and there is no enumeration of what the skill actually does beyond routing — matching the anchor 'Names the domain but actions are minimal or generic'.

2 / 5

Completeness

It states the 'when' (use only when lark-vc-agent is explicitly specified) but the 'what' is limited to 'route to lark-meeting' with no concrete capability statement; per the guideline a missing explicit 'Use when...' trigger guidance caps completeness, and the 'what' here is weak, fitting the 'only when present without a clear what' anchor.

2 / 5

Trigger Term Quality

The only natural term is the skill name itself ('lark-vc-agent'); there are no synonyms, common variations, or phrases a user would naturally say, fitting 'One or two generic keywords; missing the natural phrases users say'.

2 / 5

Distinctiveness Conflict Risk

It is narrowly scoped to the legacy name and explicitly directs to lark-meeting, giving a clear niche with minimal overlap risk; it is not a 5 only because it lacks concrete trigger phrases that fully separate it from the sibling skill.

4 / 5

Total

10

/

20

Passed

Validation

75%

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

Validation12 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 1 suspicious

Warning

Total

12

/

16

Passed

Repository
larksuite/cli
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.