Agentlas Security Scan

Agentlas Security Scan is a Quality skill for Claude Code, published by agentlas-ai in Agentlas-OS.

1.2K stars116 forkson agentlas-ai/Agentlas-OSAdded 2026/08/20Repository updated 2026/08/20

a2aagentagent-frameworkagent-os-desktopagent-skillsagentic-aiagentic-workflowagentosagentsagents-teamagentshubai-agentsautonomous-agentsllmllm-agentsllm-toolsmcpmulti-agent-systemsochestration

Install in seconds

Install Agentlas Security Scan

Copy Agentlas Security Scan into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/agentlas-ai/Agentlas-OS/tree/main/skills/agentlas-security-scan ~/.claude/skills/agentlas-security-scan

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/agentlas-ai/Agentlas-OS.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: skills/agentlas-security-scan/SKILL.md in agentlas-ai/Agentlas-OS
Installs to: ~/.claude/skills/agentlas-security-scan
Collection: One of 25 skills cataloged from this repository
Category: Quality — 1354 skills

What Agentlas Security Scan does

Agentlas Security Scan runs a two-stage security check on agent folders before private sync or public publish: a static rule scan and a BYOK LLM judgment for prompt injection, tool poisoning, secret exfiltration, destructive commands, and excessive permissions.

Agentlas Security Scan is cataloged under Quality on DirSkills. Agentlas Security Scan comes from a repository tagged a2a, agent, agent-framework, agent-os-desktop and agent-skills.

Documentation

README

Excerpt

Agentlas Security Scan (2-Stage)

Plan §6.2: stage 1 is static rule screening, stage 2 is a judgment made by the user's own LLM session (BYOK). The Cloud server never calls an LLM (v1 Non-Goal: no server-side model execution). You — the agent running this skill — are the stage-2 judge.

Stage 1 — Static scan

Run bin/hephaestus security scan <agent-folder>.
The report at .agentlas/security-scan.json lists rule-based findings ("source": "static") and a verdict: BLOCK > WARN > PASS.

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Agentlas Security Scan

What else does agentlas-ai publish alongside Agentlas Security Scan?
Agentlas Security Scan is one of 25 skills that DirSkills catalogs from agentlas-ai/Agentlas-OS, the repository it ships in. Its siblings there include Agent Team Design, Agentlas and Agentlas Auto-Activation. Each one is a separate skill with its own page in this directory, installs the same way Agentlas Security Scan does, and is maintained by agentlas-ai in that same repository. The rest of the collection is listed on the agentlas-ai/Agentlas-OS page.
How does Agentlas Security Scan compare to other Quality skills?
Agentlas Security Scan ranks #753 by stars among the 1354 Quality skills in this catalog. The most-starred ones next to it are Benchmark, Benchmark Optimization Loop and API Design Patterns. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Agentlas Security Scan against them. Open each page to compare what they document and how they install.