hackathon-doctor
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes multiple shell commands using the 'hackathon' CLI and 'git' to manage the development environment and project state. Evidence includes commands like 'hackathon doctor', 'hackathon setup', 'git config', and 'hackathon projects update'. These are central to the skill's functionality as a hackathon management tool.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, creating an attack surface where malicious input could influence agent behavior.
- Ingestion points: The agent reads and interprets JSON output from 'hackathon event status', 'hackathon doctor', and 'hackathon projects get' (found in SKILL.md).
- Boundary markers: No explicit boundary markers or 'ignore' instructions are used when interpolating CLI output into the agent's decision-making logic.
- Capability inventory: The agent has the ability to execute shell commands, modify global Git configurations, and post to chat channels via the CLI.
- Sanitization: There is no evidence of sanitization or validation of the CLI output before it is used to guide the participant through 'fixes' or pacing.
- [DATA_EXFILTRATION]: The skill sends project descriptions and requests to the hackathon platform via 'hackathon requests open' and 'hackathon chats post'. While functional, this involves transmitting local state and user-provided strings to an external API.
Audit Metadata