debrief
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
/usr/local/bin/speakbinary to provide audio feedback of the session summary. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from local project files and git logs to generate content that is subsequently passed to a system command, creating a potential injection surface.
- Ingestion points: Reads content from
tmp/handoff.md,.claude/projects/memory files, andgit logoutput. - Boundary markers: Absent; the instructions do not define delimiters to separate the external context from the agent's instructions.
- Capability inventory: Execution of shell commands via the
speakutility (skill.md). - Sanitization: Absent; while the skill strips markdown formatting, it does not perform sanitization or escaping of the content before passing it to the shell command.
Audit Metadata