SKILLEMALL.ai
GitHub ↗ RU Sign in

AB glm-search-pro

Web search via Zhipu GLM — supports both MCP (mcporter) and cURL (REST API) backends. Provides multi-engine search (Pro, Sogou, Quark, Std) with intent recognition, time range filtering, domain filtering, and configurable result count/detail level. Use when the agent needs to search the web, look up current information, find news, or retrieve online resources. Works from China without VPN. Trigger on: "search the web", "web search", "look up", "find online", "latest news", "search for", "google for", "联网搜索", "在线搜索", "查最新", "搜索一下".

ClawHub Agent Skills author: BithostGITS v1.2.5 MIT-0 5 files · 2 scripts body ≈ 1 487 tokens Open the sourceclawhub.ai↗ analyzed 3 d ago

As a process B 73/100 · Nearly there — weak spots: running it twice, progress reporting

IntegrationAI and agentsSoftware developmentInfrastructuretype 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
A
95/100
safety, quality, tests
Safety 60%
92
Quality 40%
99
Run on models
none yet
Process rating
B
73/100
Nearly there
Progress reporting w 2
0
Running it twice w 4
30
Tools and files w 18
60
the three weakest of ten parameters · all ten

What is at stake

Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.

Exfiltration medium severity

Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.

If you install

The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".

For the author

If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.

How to improve

    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
    • medium Exfiltration net-credential-use SKILL.md:141
      Credential used in a network call (verify the destination is the intended service)
      │   └── curl + $ZHIPU_API_KEY → Authorization: Bearer header → Zhipu REST API
    • low Exfiltration net-credential-use scripts/glm-search.sh:157
      Credential used in a network call (verify the destination is the intended service) (quoted — discussed, not commanded)
      elif [ -n "${ZHIPU_API_KEY:-}" ] && command -v curl &>/dev/null; then
      quoted
    • low Exfiltration net-credential-use SKILL.md:30
      Credential used in a network call (verify the destination is the intended service) (quoted — discussed, not commanded)
      No setup required. The key is read from `$ZHIPU_API_KEY` at runtime and sent via HTTP `Authorization: Bearer` header. In cURL mode, no files are written to disk.
      quoted
    • low Exfiltration net-credential-use SKILL.md:47
      Credential used in a network call (verify the destination is the intended service) (documentation table row)
      | `$ZHIPU_API_KEY` env var | Every search (cURL mode), and during setup (MCP mode) | API key |
      table

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

    Against the Agent Skills spec

    ✓ No remarks against the Agent Skills spec

    Process rating: all ten parameters 73/100

    • 0Progress reporting. Says nothing while it works
    • 30Running it twice. 1 mutating operations with no state check
    • 60Tools and files. Uses tools (web) that frontmatter does not declare
    • 60Result and completion. Output format stated, no completion criterion
    • 70When it triggers. States when to use, but not when not to
    • 70Inputs and preconditions. Inputs and preconditions are listed
    • 85Steps. 6 steps, 1 vague phrases
    • 100Failures and branches. 3 branches, has a failure section
    • 100Consistency. Name and required fields are in place
    • 100Execution cost. Instruction body is 1487 tokens
    • low 12 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)
    • +1No license
    • +2Single-language instructions
    • +5Description quotes 7 example trigger phrases
    • +3Description length 536: enough signal without eating the budget
    • +4Structure: 20 headings
    • +3Step-by-step instructions: 6 items
    • +3Output format is stated explicitly
    • +4Has examples (9 code blocks)
    • +4Reference files are cited in the instructions (1 of 1)
    • +3All 2 scripts are documented

    Quality base 70; lint remarks subtract, signals add up to 100. Result: 99.

    External checks

    ClawHub: clean
    This is a disclosed GLM/Zhipu web-search skill; its main risk is optional API-key persistence for MCP mode, not hidden or malicious behavior.
    LLM: benign (high) · VirusTotal: · 29 May 2026