ot-doctor
Pass
Audited by Gen Agent Trust Hub on Nov 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from Architecture Decision Records (ADRs) and project files to diagnose issues and suggest fixes. Maliciously crafted content in a target repository could attempt to influence the agent's behavior.
- Ingestion points: The agent reads content from
on-track/decisions/*.mdandAGENTS.mdto propose titles and identify errors (SKILL.md, steps 2 and 4). - Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the file content as untrusted data or to ignore instructions embedded within the text.
- Capability inventory: The skill can execute local shell scripts (
on-track-init.sh,on-track-index.sh), perform file writes, and move files viagit mv. - Sanitization: The skill does not perform sanitization or validation of the content extracted from project files before using it in the prompt logic.
- [COMMAND_EXECUTION]: The skill executes local bash scripts for diagnostics and uses
gitfor repository management. It also calls executable scripts from a sibling directory (../ot-init/).
Audit Metadata