north-star-investigation
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the agent to run "brutal tests," toggles, oracles, and minimal repros to validate or refute investigation hypotheses. The workflow involves executing shell commands and recording their raw results into an authoritative worklog.
- [PROMPT_INJECTION]: The skill processes external data (investigation docs, worklogs, and codebase content) while maintaining the capability to execute shell commands. This configuration creates a vulnerability surface where malicious instructions in the processed data could attempt to hijack the agent's execution loop.
- Ingestion points: Investigation documents (defined by
DOC_PATH) and files within the codebase under investigation. - Boundary markers: No specific boundary markers or "ignore" instructions are provided to delimit untrusted file content from the agent's system instructions.
- Capability inventory: The agent is authorized to perform shell command execution, file modification (worklog updates), and file reading.
- Sanitization: There are no instructions for sanitizing, escaping, or validating the content of external files before the agent processes them.
Audit Metadata