agentic-workflows

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches workflow prompts and configuration files from the github/gh-aw repository. These resources are from a well-known service provider and are essential for the skill's documented purpose of providing specialized workflow instructions.
  • [INDIRECT_PROMPT_INJECTION]: By loading and following instructions from a remote repository, the skill possesses an indirect prompt injection surface. The agent is instructed to load external Markdown files and follow them directly as new instructions.
  • Ingestion points: Multiple files located within the .github/aw/ directory of the github/gh-aw repository are loaded into the agent's context.
  • Boundary markers: None provided; the skill relies on the content of the loaded files to drive agent behavior without explicit isolation.
  • Capability inventory: The skill facilitates complex tasks including workflow creation, debugging, and optimization, which likely involve file system and command execution tools depending on the specific prompt loaded.
  • Sanitization: None; the skill assumes the integrity of the remote content hosted on the official GitHub repository.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 08:28 AM
Security Audit — agent-trust-hub — agentic-workflows