Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add InternLM/WildClawBench --skill skill-trust-auditorgit clone --depth 1 https://github.com/InternLM/WildClawBenchWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/internlm/wildclawbench/skill-trust-auditor)<a href="https://agentmods.dev/skills/internlm/wildclawbench/skill-trust-auditor"><img src="https://agentmods.dev/badge/skills/internlm/wildclawbench/skill-trust-auditor.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00019 | $0.00864 |
| Opus 5 | $0.00010 | $0.00432 |
| Sonnet 5 | $0.00004 | $0.00173 |
| Haiku 4.5 | $0.00002 | $0.00086 |
Grade A, and why
skill-trust-auditor scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
{"level": "HIGH", "pattern": "curl to external domain", "location": "scripts/sync.sh:14"}, How it starts
The opening of the file, as written. The whole thing — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Trust Auditor
Audit any ClawHub skill for security risks before installation.
🛠️ Installation
1. Ask OpenClaw (Recommended)
Tell OpenClaw: "Install the skill-trust-auditor skill." The agent will handle the installation and configuration automatically.
2. Manual Installation (CLI)
If you prefer the terminal, run:
clawhub install skill-trust-auditor
Setup (first run only)
bash scripts/setup.sh
Audit a Skill
When user says "audit [skill-name]" or "is [skill-name] safe" or before any clawhub install:
bash scripts/audit.sh [skill-name-or-url]
# Example:
bash scripts/audit.sh steipete/clawhub
bash scripts/audit.sh https://clawhub.ai/someuser/someskill
Output:
{
"skill": "someuser/someskill",
"trust_score": 72,
"verdict": "INSTALL WITH CAUTION",
"risks": [
{"level": "HIGH", "pattern": "curl to external domain", "location": "scripts/sync.sh:14"},
{"level": "MEDIUM", "pattern": "reads MEMORY.md", "location": "SKILL.md:23"}
],
"safe_patterns": ["no env var access", "no self-modification"],
"author_verified": false,
"recommendation": "Review scripts/sync.sh:14 before installing. The external curl call could exfiltrate data."
}
Post to user with clear summary:
🛡️ Trust Audit: someuser/someskill
Score: 72/100 — ⚠️ INSTALL WITH CAUTION
🔴 HIGH: curl to unknown domain in scripts/sync.sh:14
🟡 MEDIUM: reads your MEMORY.md
Recommendation: Inspect line 14 of sync.sh before proceeding.
Run: clawhub show someuser/someskill --file scripts/sync.sh
Trust Score Guide
| Score | Verdict | Action |
|---|---|---|
| 90-100 | ✅ SAFE | Install freely |
| 70-89 | ⚠️ CAUTION | Review flagged items first |
| 50-69 | 🟠 RISKY | Only if you understand the risks |
| 0-49 | 🔴 DO NOT INSTALL | High probability of malicious intent |
Risk Pattern Reference
HIGH RISK (-30 each):
process.envaccess in scriptscurl/wgetto non-standard domains- Reading
~/.configor~/.openclawdirectly exec()with user-controlled input- Instructions to modify
SOUL.md/AGENTS.md/openclaw.json
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 115 lines · 19 tokens per session scan A f9e7aa82acf6
skill-trust-auditor is a skill published in the GitHub repository InternLM/WildClawBench (516 stars, last pushed 21d ago), licensed MIT. It adds 19 tokens to every session and 864 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
agent-browser
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a…
naga-config
A configuration-management skill for Naga, including its settings, connected tools, and imported skills.
naga_control
A control interface for changing Naga’s own running state and settings through a specific agent type.
live2d_controller
// 读取：system.characterbundle.loadcharacterskillsections -> system.config.buildtier1variables.characterbuiltinskillsprompt.
code-review
A code-review and quality-analysis skill written in Chinese. It is intended to inspect code, find potential problems, and suggest improvements.
file-manager
A file and folder management helper for organizing files on a computer.