backend-to-frontend-handoff-docs
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and analyze untrusted local source code (controllers, DTOs, services) without security boundaries. Malicious instructions placed in code comments or string literals within a repository could attempt to hijack the agent's execution context during the documentation generation process.
- Ingestion points: Processes backend source code files including controllers, DTOs, and validation logic as defined in the Workflow section of SKILL.md.
- Boundary markers: Absent. The instructions do not provide delimiters or "ignore embedded instructions" warnings for the agent when reading external files.
- Capability inventory: The agent is instructed to perform file system write operations to
.claude/docs/ai/(SKILL.md). - Sanitization: Absent. There is no requirement for the agent to filter or sanitize the extracted content before writing it to the handoff files.
- [PROMPT_INJECTION]: The skill uses instructions to suppress user interaction, such as 'No Chat Output', 'No discussion', and 'no explanation' (SKILL.md). This concealment pattern reduces human-in-the-loop oversight by preventing the agent from explaining its actions or the content it is generating, which can be leveraged to hide malicious output if the agent is successfully injected.
Audit Metadata