hermes-atlas-ecosystem-map

Pass

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires downloading project source code from an external GitHub repository: https://github.com/ksimback/hermes-ecosystem.git.
  • [INDIRECT_PROMPT_INJECTION]: The RAG chatbot retrieves information from data/repos.json, a community-curated file. Malicious data in this file could potentially manipulate the chatbot's outputs.
  • Ingestion points: data/repos.json and markdown files in the research/ directory.
  • Boundary markers: The retrieval pipeline lacks explicit delimiters or instructions to prevent the agent from following commands embedded in retrieved chunks.
  • Capability inventory: The skill uses network operations to interact with OpenRouter, Redis, and the GitHub API.
  • Sanitization: No validation or filtering is described for the community-contributed data before it is indexed or retrieved.
  • [COMMAND_EXECUTION]: The instructions involve running several local commands and scripts, such as npm install, node scripts/build-chunks.js, and vercel dev, which execute code with the user's system privileges.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 12, 2026, 06:50 PM
Security Audit — agent-trust-hub — hermes-atlas-ecosystem-map