CtrlK
BlogDocsLog inGet started
Tessl Logo

knowledge

本地知识库集成 - 文档检索、投喂、双轨模式切换

48

Quality

50%

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 ./configs/microservice/bff-service/configs/agent-skills/clawhub/knowledge/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

67%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.

The body is a compact, well-organized command reference with concrete triggers and endpoints that suits a simple skill. Its main weakness is actionability: it documents the API endpoint but gives no example request, so Claude cannot execute retrieval directly without inference.

Suggestions

Add one or two executable example requests against http://127.0.0.1:8001 (e.g. a search/curl call) so the retrieval action is copy-paste ready.

Include a brief confirmation or validation cue after mode switches (e.g. how to verify which track is active) to lift workflow clarity.

Clarify what "投喂" (ingestion) entails concretely — a one-line command or API call — since it is listed as a capability but has no corresponding row in the command table.

DimensionReasoningScore

Conciseness

The body is a tight 16-line table of commands plus a short file-location list with no padding or explanations of concepts Claude already knows, fitting the "efficient; minor instances that could be trimmed" anchor; it does not quite reach fully lean because the command table could be marginally denser.

4 / 5

Actionability

It provides concrete voice-command strings and exact file paths/URLs, but gives no executable API calls or request examples for the documented endpoints, so guidance is concrete yet incomplete for actually performing retrieval.

3 / 5

Workflow Clarity

The command table cleanly maps each trigger to its action and the file-location section gives concrete endpoints, yielding a clear single-purpose sequence; it stops short of 5 because there are no explicit validation or confirmation steps after a retrieval/mode switch.

4 / 5

Progressive Disclosure

Content is appropriately organized into a commands table and a file-location section with no inlined bulk reference material and no nested references; as a short single-purpose skill with no bundle files this is well-structured, though the sections are unadorned rather than explicitly signaled for navigation.

4 / 5

Total

15

/

20

Passed

Description

33%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 states the skill's domain and a few concrete functions but omits any explicit invocation trigger ("when to use"), which is the most consequential gap. Trigger terms are generic rather than the natural phrasings users would say.

Suggestions

Add an explicit "Use when..." clause naming concrete user triggers (e.g. "Use when the user says '查一下', asks to search the local knowledge base, or wants to switch retrieval modes.").

Surface the natural voice-command triggers ("查一下 xxx", "切换到本地知识库") directly in the description so they act as discoverable keywords.

Distinguish the skill from general RAG/search by naming the specific local API endpoint or "AnythingLLM 双轨" context in the description.

DimensionReasoningScore

Specificity

The description names the domain ("本地知识库集成") and 2-3 concrete capabilities ("文档检索、投喂、双轨模式切换"), matching the anchor that lists domain plus 1-2 concrete actions but is not comprehensive.

3 / 5

Completeness

It gives a clear "what" but no "when" clause — there is no "Use when..." guidance telling Claude when to invoke this skill, capping completeness well below a passing level.

2 / 5

Trigger Term Quality

It uses a few generic terms ("知识库", "检索", "投喂") but lacks the natural trigger phrases a user would actually say such as quoted voice commands found only in the body; coverage is minimal and missing common variations.

2 / 5

Distinctiveness Conflict Risk

"本地知识库集成" is somewhat specific and anchored to a local knowledge base, reducing conflict risk, yet the broad retrieval/ingestion framing could still overlap with general search or RAG skills.

3 / 5

Total

10

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
UnicomAI/wanwu
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.