ClawHub
Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Auto Skill Distiller

v1.0.0

Auto-distill successful workflows into reusable skills. Use after completing any multi-step task to evaluate if the workflow should be saved as a skill. Trig...

⭐ 0· 49·0 current·0 all-time
by@christianye
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description match the runtime instructions: the document describes how to extract workflows, generalize them, and save SKILL.md files under ~/.openclaw/skills. References to integration (trinity-harness Layer 3) are optional integrations, not hidden requirements.
✓
Instruction Scope
SKILL.md instructs the agent to inspect recent workflow steps, generalize them, run quality checks, and write files to ~/.openclaw/skills/<slug>/SKILL.md (and optional references/). It also suggests using commands like ls and read against the skills directory. There are no instructions to read unrelated system files, access credentials, or send data to external endpoints.
✓
Install Mechanism
Instruction-only skill with no install spec, no code to download, and no required binaries — lowest-risk install surface.
✓
Credentials
No environment variables, credentials, or config paths are required. The only filesystem access described is under the user's ~/.openclaw/skills and memory files, which is proportional to the stated purpose.
ℹ
Persistence & Privilege
The skill directs writing files into ~/.openclaw/skills (its expected scope). It does not set always:true. However, if your agent/platform enables autonomous triggers (Layer 3 compound mode), the agent could create SKILL.md files automatically — the doc advises announcing creations, but you should confirm the platform enforces review before committing changes.
Assessment
This instruction-only skill is coherent and low-risk: it only creates/edits SKILL.md under ~/.openclaw/skills and performs local checks. Before installing or enabling automatic distillation: (1) ensure you want the agent to be able to write into ~/.openclaw/skills (backup that directory if needed); (2) confirm your platform enforces a review/approval step so created skills aren’t added silently (the SKILL.md itself says to announce creations); (3) review any generated SKILL.md for accidental inclusion of sensitive conversation content before saving; and (4) if you do not want autonomous creation, keep autonomous invocation constrained or disable any Layer 3 automatic distillation integration. Overall this skill appears to do what it claims.

Like a lobster shell, security has layers — review code before you run it.

automationvk9777pcmxznf1gvk7cpcdyzc7584hb7fcompoundvk9777pcmxznf1gvk7cpcdyzc7584hb7fdistillvk9777pcmxznf1gvk7cpcdyzc7584hb7flatestvk9777pcmxznf1gvk7cpcdyzc7584hb7flearningvk9777pcmxznf1gvk7cpcdyzc7584hb7fskillvk9777pcmxznf1gvk7cpcdyzc7584hb7fworkflowvk9777pcmxznf1gvk7cpcdyzc7584hb7f

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments