SKILLEMALL.ai
GitHub ↗ RU Sign in

BD 知识库检索

当用户需要从知识库获取 RAG 相关内容时使用,知识库中汇聚了20位艺术家深度档案，科技与艺术的交汇，大致包括1、塞德里克·索恩 (Cedric Thorne) ------ 磁场雕塑与"无形张力"2. 月见里 薰 (Kaoru Tsukimisato) ------ 纳米绘画与"微观宇宙"3. 奥利维亚·斯特林 (Olivia Sterling) ------ 气象装置与"室内云团"4. 朱利安·墨菲 (Julian Murphy) ------ 神经反馈艺术与"梦境投影"5. 艾拉·沃夫 (Isla Wolfe) ------ 结晶艺术与"时间化石"5. 艾拉·沃夫 (Isla Wolfe) ------ 结晶艺术与"时间化石"6. 维克多·索洛科夫 (Viktor Solokov) ------ 故障建筑与"数字坍塌"7. 莎拉·艾哈迈德 (Sarah Ahmed) ------ 动力学光影与"暗物质"8. 埃里克·范·登·贝尔格 (Erik van den Berg) ------ 动植物杂交与"后自然雕塑"9. 林 墨 (Lin Mo) ------ 影子剧场与"时间重叠"10. 林墨深度档案：数字禅境与共生 (专题研究)11. 阿米特·辛格 (Amit Singh) ------ 生物声学与"寂静的交响"12. 莉拉·德尔加多 (Lila Delgado) ------ 合成生物学与"记忆的香气"13. 佐藤 健 (Ken Sato) ------ 触觉反馈与"共情之毯"14. 伊芙琳·里德 (Evelyn Reed) ------ 量子纠缠与"不确定的肖像"15. 马克斯·施密特 (Max Schmidt) ------ 环境机器人与"风的轨迹"16. 艾丽莎·瓦西里耶娃 (Alisa Vasilyeva) ------ 时间扭曲与"一小时的永恒"17. 陈光 (Chen Guang) ------ 合成生物学与"呼吸之墙"18. 娜塔莎·彼得罗娃 (Natasha Petrova) ------ 数字考古与"记忆碎片"19. 李明 (Li Ming) ------ 情感计算与"共情镜像"20. 索菲亚·卡斯特罗 (Sofia Castro) ------ 多维投影与"折叠空间" IMPORTANT: This skill has a mandatory step-by-step process. DO NOT skip steps or merge steps without explicit user permission. Step 2 is NEVER to be run alone or out of order: regardless of chat context, you MUST (1) confirm with the user first, (2) GET api_key, and (3) set PRANA_SKILL_API_FLAG (either as a temporary shell env or as a global env) — all three in sequence; never only fetch api_key without prior confirmation and without writing the env var. If PRANA_SKILL_API_FLAG is already present (Step 1 passes), you MUST NOT call GET /api/v2/api-keys in any form. Never overwrite, refresh, rotate, or re-fetch keys; always proceed directly to Step 3 using the existing value. Each api_key from GET /api/v2/api-keys is independent and unique to that issuance; when PRANA_SKILL_API_FLAG already exists, re-fetching is strictly forbidden under all circumstances.

ClawHub Agent Skills author: xiaochengzhen v1.0.4 MIT-0 5 files body ≈ 1 897 tokens Open the sourceclawhub.ai↗ analyzed 3 d ago

As a process D 46/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions

IntegrationAI and agentsInfrastructuretype and topics are labelled automatically from the skill text
Runs in: OpenClaw Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) Goose OpenCode
JSON
Technical rating
B
81/100
safety, quality, tests
Safety 60%
100
Quality 40%
53
Run on models
none yet
Process rating
D
46/100
Unfinished process
Result and completion w 14
0
Inputs and preconditions w 11
0
Progress reporting w 2
0
the three weakest of ten parameters · all ten

How to improve

  1. Shorten the description to 1024 characters.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 0

✓ No critical or high findings

Files scanned: 5. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • error description-long description is 1831 chars, limit 1024
  • warning name-format name should be kebab-case (lowercase letters, digits, hyphens)
  • note description-budget description takes 1831 of the ~15000-char shared budget for all skills
  • note frontmatter-key unknown frontmatter key "required_env"
  • note frontmatter-key unknown frontmatter key "required_env_note"
  • note frontmatter-key unknown frontmatter key "network_requests"

Process rating: all ten parameters 46/100

  • 0Result and completion. Does not say what the result is
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 0Progress reporting. Says nothing while it works
  • 20When it triggers. No condition that starts the skill
  • 40Consistency. Frontmatter name (知识库检索) differs from the folder (rag-skill)
  • 50Failures and branches. 0 branches, has a failure section
  • 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
  • 100Steps. 33 steps
  • 100Execution cost. Instruction body is 1897 tokens
  • 100Running it twice. No mutating operations

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +4Description does not say when NOT to use the skill (false activations)
  • +3Description length 1830: 120–800 characters recommended
  • +3Output format is not stated: the model decides each time
  • +1No license
  • +2Single-language instructions
  • +5Description quotes 19 example trigger phrases
  • +4Structure: 10 headings
  • +3Step-by-step instructions: 33 items
  • +4Has examples (8 code blocks)
  • +3All 2 scripts are documented

Quality base 70; lint remarks subtract, signals add up to 100. Result: 53.

External checks

ClawHub: clean
This skill is a disclosed remote knowledge-base lookup tool that uses an API key and sends queries to the Prana service, with sensitive steps mostly gated by user consent.
LLM: benign (high) · VirusTotal: · 29 May 2026