agentic-awesome-skills@19.8.0
⚠ Under reviewAAS Core: complete local skill discovery, agent-owned selection, stack validation, and planning, backed by 2,705+ agentic skills.
AI Security Review
scanned 5d ago · by lpm-firewall-aiThe inspected scripts and workflows show user-directed functionality and no confirmed attack. The required PDF artifact remains structurally uninspected, so the evidence does not support either a clean verdict or a concrete warning.
Decision evidence
public snapshot- The Apify analytics workflow asks for an actor and user preferences before running its script; the script reads APIFY_TOKEN and sends requests to api.apify.com.
- The installer registers agent skill directories as targets; the inspected code does not establish an unconsented lifecycle write.
- The theme skill describes theme-showcase.pdf as a visual reference to display, but the artifact could not be structurally inspected, so its contents remain unresolved.
- The Apify runner requires explicit actor and input arguments and uses the Apify API for its actor requests.
- The Hugging Face example sends HF_TOKEN only as authorization to huggingface.co; the Fedora shell file is an isolated, temporary-home test runner.
Source & flagged code
13 flagged · loading sourcePackage contains a high-severity secret pattern.
skills/developer-signup-flow/SKILL.mdView on unpkg · L165Stripe test secret key in skills/developer-signup-flow/SKILL.md
skills/developer-signup-flow/SKILL.mdView on unpkg · L165Package source references child process execution.
tools/bin/install.jsView on unpkg · L2Findings
Affected versions and remediation
This report applies to agentic-awesome-skills@19.8.0.
See all versions of agentic-awesome-skills for other recorded verdicts.
- Review the evidence and your use of agentic-awesome-skills@19.8.0 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.