SKILLEMALL.ai
GitHub ↗ RU Sign in

BF research-pro

系统化研究 skill — 螺旋收敛模型。把任何问题（模糊或清晰）分解成子问题，迭代搜索，越搜越清晰，直到每个子问题都有答案。 Triggers: "帮我研究", "研究一下", "调研", "分析对比", "research", "investigate", "look up" 也触发: 竞品分析、市场调研、技术选型对比、趋势了解 Setup triggers: "安装 research-pro", "配置 research-pro", "research-pro doctor", "setup research-pro", "research-pro 未就绪" **Gates:** 开始外部搜索前必须通过 READY doctor 与 Search Target Confirmation Gate；所有交互式多源/外部研究必须先展示 Search Contract 并收到用户明确确认。唯一例外是一个用户明确提供的 URL/文件、且只要求读取/提取/摘要、不作跨源比较或推断的 `NARROW_SELFCHECK`（见 SKILL Phase 1）。 Does NOT trigger normal multi-source research: - 已经知道答案的简单事实问题 - 用户直接给了 URL/文件、且只要读取/提取/摘要的，按 Step 1.2 的 `NARROW_SELFCHECK` 处理（不进入跨源 research） - 代码调试、写代码任务 Output: 结构化研究报告（结论 + 子问题答案 + 来源 + 争议点 + 未解决缺口）

ClawHub Hermes author: Ming Fang v3.17.2 MIT-0 39 files · 3 scripts body ≈ 6 521 tokens Open the sourceclawhub.ai↗ analyzed 3 d ago

系统化研究 skill — 螺旋收敛模型。把任何问题（模糊或清晰）分解成子问题，迭代搜索，越搜越清晰，直到每个子问题都有答案。 Triggers: "帮我研究", "研究一下", "调研", "分析对比", "research", "investigate", "look up" 也触发…

As a process F 38/100 · Will not run — References files that are not bundled: URL

IntegrationYouTubeMedia and videotype and topics are labelled automatically from the skill text
Runs in: OpenClaw Hermes Agent
JSON
Technical rating
B
84/100
safety, quality, tests
Safety 60%
96
Quality 40%
66
Run on models
none yet
Process rating
F
38/100
Will not run
References files that are not bundled: URL
Tools and files w 18
0
Result and completion w 14
0
Inputs and preconditions w 11
0
the three weakest of ten parameters · all ten

How to improve

  1. For Hermes the description must be one sentence under 60 characters; move the conditions to a "When to Use" section.
  2. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
  3. The text references files that are not there: add them or drop the references.
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 · 4

✓ No critical or high findings

Medium and low: 4
  • low Exfiltration net-credential-use SKILL.md:373
    Credential used in a network call (verify the destination is the intended service) (quoted — discussed, not commanded)
    > ⚠️ 下面凡是裸 CLI（tvly/firecrawl/youtube_transcript_api）或用到 `$XXX_API_KEY` 的 curl，**都要经凭据 shim**，否则冷启动报 "No API key"。
    quoted
  • low Exfiltration exfil-secret-in-url SKILL.md:396
    Credential passed in a URL query string (normal for some APIs — verify the host is the intended service) (destination is a well-known publishing service; quoted — discussed, not commanded)
    RWC bash -c 'curl "https://www.googleapis.com/youtube/v3/search?part=…&q=QUERY&type=video&maxResults=5&key=…"'
    known servicequoted
  • low Exfiltration net-credential-use SKILL.md:396
    Credential used in a network call (verify the destination is the intended service) (destination is a well-known publishing service; quoted — discussed, not commanded)
    RWC bash -c 'curl "https://www.googleapis.com/youtube/v3/search?part=…&q=QUERY&type=video&maxResults=5&key=…"'
    known servicequoted
  • low Exfiltration read-dotenv SKILL.md:428
    Reads a .env file
    # Credentials: scripts auto-load via lib/credentials.mjs — do NOT hardcode source ~/.openclaw/.env

Files scanned: 37. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning description-long-hermes description is 690 chars; the Hermes authoring standard requires ≤ 60 (one sentence, ending with a period)
  • warning body-long SKILL.md body ≈ 6521 tokens (recommended < 5000); move details to references/
  • warning missing-ref reference to a missing file: URL

Process rating: all ten parameters 38/100

Will not run. References files that are not bundled: URL
  • 0Tools and files. 1 referenced file(s) missing: URL
  • 0Result and completion. Does not say what the result is
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 30Running it twice. 1 mutating operations with no state check
  • 50When it triggers. No condition that starts the skill
  • 50Failures and branches. 0 branches, has a failure section
  • 60Consistency. The Hermes dialect needs category and tags
  • 70Execution cost. Instruction body is 6521 tokens
  • 100Steps. 155 steps
  • 100Progress reporting. Reports progress
  • low 13 top-level sections: this looks like several domains in one skill

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)
  • +3Output format is not stated: the model decides each time
  • -247 emoji in the instructions: noise for the model
  • -35 of 13 scripts are never mentioned in SKILL.md
  • +1No license
  • +2Single-language instructions
  • +5Description quotes 8 example trigger phrases
  • +3Description length 689: enough signal without eating the budget
  • +4Structure: 32 headings
  • +3Step-by-step instructions: 155 items
  • +4Has examples (26 code blocks)
  • +4Reference files are cited in the instructions (1 of 6)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 66.

External checks

ClawHub: suspicious
This is a real research skill, but it grants and reuses credentials, modifies multiple agent environments, performs external calls, and keeps research traces in ways users should review before installing.
LLM: suspicious (high) · 30 Aug 2026