cco-overhead
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Node.js script located at
${CLAUDE_PLUGIN_ROOT}/src/overhead.js. It also facilitates the execution of CLI commands likeclaude mcp removeto modify the agent's environment based on the audit results. - [INDIRECT_PROMPT_INJECTION]: The skill processes session transcripts, which contain untrusted data from user inputs and tool outputs, to calculate token usage and overhead.
- Ingestion points: Processes session transcripts and project-specific files like
CLAUDE.mdto measure context size. - Boundary markers: No specific delimiters or safety instructions are defined for the transcript parsing process in the documentation.
- Capability inventory: The skill utilizes the Bash tool to execute scripts and environment modification commands.
- Sanitization: The instructions explicitly forbid the agent from constructing removal commands that were not printed by the audit report, relying on the integrity of the underlying script.
Audit Metadata