release-director

Pass

Audited by Gen Agent Trust Hub on Mar 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill implements an indirect prompt injection surface through the loading of external configuration and metadata.
  • Ingestion points: The agent is instructed to call load_override('release-preferences.md') to incorporate custom user instructions and reads metadata from the album README file in the pre-release phase.
  • Boundary markers: There are no explicit instructions to use delimiters or ignore embedded natural language instructions when processing the contents of the override file or the album metadata.
  • Capability inventory: The skill has access to powerful tools including Bash, Edit, Write, and several MCP tools (check_streaming_lyrics, update_streaming_url, verify_streaming_urls) which could be misused if the agent is influenced by malicious instructions in the input data.
  • Sanitization: The skill lacks explicit validation or escaping requirements for the dynamically processed content, such as the album name used in social media templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 27, 2026, 11:26 PM
Security Audit — agent-trust-hub — release-director