sprint-forge
Pass
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its core workflow of analyzing untrusted codebases. \n
- Ingestion points: The 'Deep Analysis' phase in
assets/modes/INIT.mdand the task execution loop inassets/modes/SPRINT.mdread arbitrary content from the project directory. \n - Boundary markers: Absent; the prompts do not provide instructions to differentiate between project code and instructions, allowing malicious comments in code to be interpreted as tasks. \n
- Capability inventory: The skill leverages
Bash,Write,Edit, andTasktools to modify files and run system commands. \n - Sanitization: Absent; codebase contents are interpolated directly into the context for decision-making and task generation.\n- [COMMAND_EXECUTION]: The SPRINT execution workflow (
assets/modes/SPRINT.md, Step 8) allows the agent to run verification commands and apply changes using theBashtool. These commands are derived from tasks generated during the analysis of the user's local files.
Audit Metadata