minions-hermes-mission-control

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill utilizes npx to download and execute the minionsai package directly from the NPM registry. This is a standard installation method for the vendor's tool.
  • [COMMAND_EXECUTION]: The instructions include shell commands for starting the server, checking version numbers, and managing the local SQLite database files (e.g., rm -rf ~/.minions/).
  • [PROMPT_INJECTION]: As a controller for autonomous agents, the skill exposes a surface for indirect prompt injection through its core functionality of processing untrusted data.
  • Ingestion points: The skill explicitly documents tasks that scrape external websites (e.g., competitor.com, HackerNews) and read local file content (e.g., content/drafts/).
  • Boundary markers: No specific delimiters or "ignore instructions" wrappers are mentioned in the prompt examples provided in SKILL.md to isolate untrusted data.
  • Capability inventory: The managed agents have extensive capabilities, including executing parallel network requests, writing files to the local workspace, and autonomous reasoning loops (SKILL.md).
  • Sanitization: The documentation does not specify sanitization or validation routines for data ingested from external sources before it is processed by the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 11:19 PM
Security Audit — agent-trust-hub — minions-hermes-mission-control