Agent Skill Trust Check
Use this skill before installing a third-party agent skill, SKILL.md package, MCP-linked skill, or marketplace listing.
What This Skill Does
- Reads a public or local skill description before install.
- Flags patterns that deserve review: shell execution, destructive commands, secrets, wallet/payment actions, network output, persistence, and prompt-boundary issues.
- Separates basic provenance signals from risky behavior signals.
- Produces a patch order that a maintainer can resolve before the skill is trusted.
When To Use
Use this when an agent is about to install or recommend a skill from:
- OpenClaw, Hermes, or ClawHub-style marketplaces.
- Claude Code, Codex, Cursor, Windsurf, or Gemini skill directories.
- MCP-linked skill bundles.
- GitHub repositories that include a
SKILL.md, tool manifest, or install script.
Run From The Public Repo
git clone https://github.com/TateLyman/agent-skill-trust-check.git
cd agent-skill-trust-check
npm run check
node bin/agent-skill-trust-check.js ./SKILL.md
Marketplace-safe stdin mode:
node bin/agent-skill-trust-check-stdin.js < ./SKILL.md
For JSON output:
node bin/agent-skill-trust-check.js ./SKILL.md --json
npm command:
npx --yes agent-skill-trust-check ./SKILL.md --json
Review Rules
Before installation, check:
- Does the skill run shell commands, package installers, or process-spawn APIs?
- Does it read secrets, environment variables, wallet data, credentials, cookies, or private keys?
- Does it send content to remote URLs or webhooks?
- Does it create persistent background jobs?
- Does it ask the agent to ignore or override higher-priority instructions?
- Does it document source, license, version, tests, permissions, and uninstall steps?
Boundaries
This is a static pre-install check. The marketplace-safe runner reads only stdin and returns JSON. The local CLI can also read a local path or a public GitHub/raw/Gist URL when run from the repository checkout. Neither mode executes the target skill or proves the runtime is safe.
For marketplace-grade review, use the paid Agent Skill Trust Check listing:
https://orkai.ai/skills/agent-skill-trust-check
1---
2spec: usk/1.0
3name: agent-skill-trust-check
4version: 0.1.2
5description: Static pre-install trust review for SKILL.md, OpenClaw, Hermes, MCP, and agent-skill marketplace packages before they request local, account, payment, or external access.
6interface:
7 type: cli
8 entry_point: bin/agent-skill-trust-check-stdin.js
9 runtime: node
10 call_pattern: stdin_stdout
11input_schema:
12 type: object
13 properties:
14 text:
15 type: string
16 description: SKILL.md, marketplace description, or install instructions to review before installation.
17 target:
18 type: string
19 description: Optional source label for the reviewed text.
20 required:
21 - text
22output_schema:
23 type: object
24 properties:
25 verdict:
26 type: string
27 description: Static install-readiness verdict.
28 risk_score:
29 type: integer
30 description: Aggregate static risk score.
31 findings:
32 type: array
33 description: Matched risky behavior signals and suggested fixes.
34 missing_signals:
35 type: array
36 description: Provenance or safety signals not found in the target text.
37permissions:
38 network: false
39 filesystem: false
40 subprocess: false
41 env_vars: []
42category: security
43capabilities:
44 - skill_review
45 - agent_skill_security
46 - marketplace_vetting
47 - mcp_review
48license: MIT
49tags:
50 - security
51 - agent-skills
52 - skill-review
53 - mcp
54author: Tate Programs
55homepage: https://github.com/TateLyman/agent-skill-trust-check
56platform_compatibility:
57 - Claude Code
58 - Codex CLI
59 - Cursor
60 - Gemini CLI
61 - OpenClaw Agent
62requirements:
63 node: ">=20"
64changelog: Added a marketplace-safe stdin runner that needs no filesystem or network access.
65---
66
67# Agent Skill Trust Check
68
69Use this skill before installing a third-party agent skill, SKILL.md package, MCP-linked skill, or marketplace listing.
70
71## What This Skill Does
72
73- Reads a public or local skill description before install.
74- Flags patterns that deserve review: shell execution, destructive commands, secrets, wallet/payment actions, network output, persistence, and prompt-boundary issues.
75- Separates basic provenance signals from risky behavior signals.
76- Produces a patch order that a maintainer can resolve before the skill is trusted.
77
78## When To Use
79
80Use this when an agent is about to install or recommend a skill from:
81
82- OpenClaw, Hermes, or ClawHub-style marketplaces.
83- Claude Code, Codex, Cursor, Windsurf, or Gemini skill directories.
84- MCP-linked skill bundles.
85- GitHub repositories that include a `SKILL.md`, tool manifest, or install script.
86
87## Run From The Public Repo
88
89```bash
90git clone https://github.com/TateLyman/agent-skill-trust-check.git
91cd agent-skill-trust-check
92npm run check
93node bin/agent-skill-trust-check.js ./SKILL.md
94```
95
96Marketplace-safe stdin mode:
97
98```bash
99node bin/agent-skill-trust-check-stdin.js < ./SKILL.md
100```
101
102For JSON output:
103
104```bash
105node bin/agent-skill-trust-check.js ./SKILL.md --json
106```
107
108npm command:
109
110```bash
111npx --yes agent-skill-trust-check ./SKILL.md --json
112```
113
114## Review Rules
115
116Before installation, check:
117
118- Does the skill run shell commands, package installers, or process-spawn APIs?
119- Does it read secrets, environment variables, wallet data, credentials, cookies, or private keys?
120- Does it send content to remote URLs or webhooks?
121- Does it create persistent background jobs?
122- Does it ask the agent to ignore or override higher-priority instructions?
123- Does it document source, license, version, tests, permissions, and uninstall steps?
124
125## Boundaries
126
127This is a static pre-install check. The marketplace-safe runner reads only stdin and returns JSON. The local CLI can also read a local path or a public GitHub/raw/Gist URL when run from the repository checkout. Neither mode executes the target skill or proves the runtime is safe.
128
129For marketplace-grade review, use the paid Agent Skill Trust Check listing:
130
131https://orkai.ai/skills/agent-skill-trust-check
aiskillstore