debug-workflow
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill defines a protocol for processing external data such as bug reports, error logs, and source code. This processing of untrusted content creates a vulnerability to indirect prompt injection, where malicious instructions could be embedded in the data provided to the agent.
- Ingestion points: Bug reports, error messages, and source code files mentioned throughout the Reproduction and Location phases of SKILL.md.
- Boundary markers: The skill lacks explicit instructions for the agent to use delimiters or ignore instructions found within the untrusted debugging data.
- Capability inventory: The agent is instructed to reproduce bugs (execute code) and apply fixes (write files) across all phases of the workflow.
- Sanitization: No sanitization or validation of the input data is described in the workflow.
Audit Metadata