plan-implement-verify-recover
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process codebase content (source code, tests, and instructions) to perform modifications, which creates a vulnerability surface for indirect prompt injection if that code contains malicious instructions.
- Ingestion points: The agent is instructed to read implementation files, callers, tests, and repository-specific instructions (
AGENTS.md) in thediscoverandStandard/Deeproutes (SKILL.md). - Boundary markers: The skill does not define explicit delimiters or instructions to ignore embedded prompts within the files being read.
- Capability inventory: The agent has the authority to read files, write changes to the repository, and execute project-defined build, test, and lint commands.
- Sanitization: The workflow does not include specific sanitization or validation steps for content ingested from the external repository files before processing them.
Audit Metadata