skill-doc-delivery
Pass
Audited by Gen Agent Trust Hub on Mar 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill performs legitimate document management tasks using well-known services and standard commands.
- [COMMAND_EXECUTION]: The skill uses standard system commands (ls, cat) to identify and read markdown files for conversion. These operations are restricted to the user's local output directory (~/.claude-octopus/results/) and align with the skill's stated purpose.
- [EXTERNAL_DOWNLOADS]: Instructions are provided to install a document-skills plugin from the anthropic-agent-skills source. This reference is associated with a well-known technology provider and follows standard extension practices.
- [PROMPT_INJECTION]: The skill uses metadata triggers (e.g., "AUTOMATICALLY ACTIVATE when user requests") to facilitate intent recognition for document export. These instructions are standard functional descriptors and do not attempt to override agent safety protocols.
Audit Metadata