LYGO Skill Gate — local pre-install skill risk scanner for OpenClaw/ClawHub packages. Scan any skill folder before you install or trust it: subprocess/shell, network/HTTP, secrets in source, eval/exec, webhook/exfil hints, permission-claim mismatches. Use when auditing ClawHub skills, reviewing SKILL.md safety, safe install checks, malware triage, or SkillSpector-style local gates. Pure stdlib. No network, no subprocess, no auto-install. Install clawhub:@deepseekoracle/lygo-skill-gate.
Documents
LYGO SkillSpector
Try itLYGO SkillSpector — enhanced local pre-install skill risk scanner for OpenClaw/ClawHub packages. Scan, gate (CI max-band), batch, and Markdown report: subprocess/shell, network/HTTP/httpx, secrets, eval/exec, curl|bash, miners, claim mismatches. Use when auditing ClawHub skills, SkillSpector-style gates, safe install checks, malware triage. Pure stdlib. No network, no subprocess, no auto-install. Builder HTML/multi-root pack on SkillHub FULL if running a full stack. Install clawhub:@deepseekoracle/lygo-skill-spector.
What it does
LYGO SkillSpector — enhanced local pre-install skill risk scanner for OpenClaw/ClawHub packages. Scan, gate (CI max-band), batch, and Markdown report: subprocess/shell, network/HTTP/httpx, secrets, eval/exec, curl|bash, miners, claim mismatches. Use when auditing ClawHub skills, SkillSpector-style gates, safe install checks, malware triage. Pure stdlib. No network, no subprocess, no auto-install. Builder HTML/multi-root pack on SkillHub FULL if running a full stack. Install clawhub:@deepseekoracle/lygo-skill-spector.
The skill document
LYGO SkillSpector v1.0.1
Scan before you install — gate before you ship.
Enhanced local risk scanner for OpenClaw/ClawHub skill packages. Successor surface to Skill Gate with CI gate, batch, Markdown report, and deeper rule pack (httpx, curl|bash, PowerShell IEX, miners, HF/OpenAI key shapes).
Signature: Delta9Phi963-SKILL-SPECTOR-v1.0.1
ClawHub: @deepseekoracle/lygo-skill-spector
Audit response: references/SKILLSPECTOR_AUDIT.md (v1.0.0 meta-scan false positives fixed)
Builder version (full stack): If you run a full LYGO stack, the unlocked builder pack (HTML multi-root reports, multi-gate CI matrix, dashboard JSON) is on
SkillHub FULL LYGO
aslygo-skill-spectorFULL RAW (+builder/).
Dual channel
| Channel | What you get |
|---|---|
| ClawHub (public) | scan · gate · batch · report · self-demo — tentacle |
| SkillHub FULL | Same + builder/ HTML batch, multi-root gate, CI summary — unlocked RAW |
Install (public)
npx clawhub@latest install deepseekoracle/lygo-skill-spector
FULL engineer RAW: https://chatagent.ca/lygoskillhub.html#full-lygo
Commands
cd path/to/lygo-skill-spector
python scripts/self_check.py
# Scan one skill
python scripts/skill_spector.py scan "I:/E Drive/.grok/skills/lygo-context-guard"
python scripts/skill_spector.py scan lygo-ops-detector
# CI gate (exit non-zero if risk worse than max-band)
python scripts/skill_spector.py gate ./some-skill --max-band low
# Batch under a skills root
python scripts/skill_spector.py batch "I:/E Drive/.grok/skills"
# Markdown report
python scripts/skill_spector.py report ./some-skill
python scripts/skill_spector.py report ./some-skill --write last.md --i-consent
# Self-demo
python scripts/skill_spector.py self-demo
FULL builder only (SkillHub)
python builder/skill_spector_builder.py html-batch ./skills --write batch.html --i-consent
python builder/skill_spector_builder.py multi-gate ./skills --max-band elevated
python builder/skill_spector_builder.py ci-summary ./skills --write ci.json --i-consent
| Exit | Meaning |
|---|---|
| 0 | clear / low (or under max-band) |
| 5 | elevated or claim mismatch |
| 10 | high / critical |
| 2 | bad path / not a skill |
What it checks
subprocess/os.system/Popen/shell=True/ PowerShellIEX- HTTP clients (
urllib,requests,httpx,aiohttp), sockets, webhooks curl|bash/wget|bashremote-code patternseval/exec, pickle, unsafe yaml- Hardcoded key-like / token-shaped strings (vendor project-key and HF-style prefixes)
- Destructive deletes, force-push, auto-publish / ClawHub publish hints
- Mining / keylogger-style IOC detection signals (rules only — not mining code)
- Claim mismatch: frontmatter says
network: falsebut code uses HTTP
Does not: download skills, install skills, execute scanned code, or phone home.
Pair with
| Skill | Role |
|---|---|
lygo-skill-gate | Lighter single-scan gate (still supported) |
lygo-context-guard | Token budget + secret redact |
lygo-continuum | Seal “scanned clean” as checkable claims |
lygo-kickstart-wizard | Onboarding map |
License
MIT-0.
Δ9Φ963 — verify before trust · local spector · human remains the publisher.
Related skills
Automated security audit for AI agent skills. Use BEFORE installing any skill from ClawHub, GitHub, or other sources. Scans SKILL.md + all files for 30+ red...
by Zoran2 installs
Get a PASS/WARN/BLOCK verdict on a ClawHub skill before installing it, with evidence for any findings.
21 installs
ClawHub reputation checker for clawsec-suite. Adds a standalone reputation gate before guarded skill installation.
34 installs
LYGO CLI Bridge — unified entrypoint: lygo health | map | analyze | mint | radar | next. Wraps kickstart + ops-detector + mint-walkthrough + deception-radar without exposing internal skill layout. In-process imports only; no subprocess; no auto-publish. Install clawhub:@deepseekoracle/lygo-cli-bridge.
1 installs
Deep audit for installed ClawHub skills — usage analysis, permission review, conflict detection
4 installs