wjs-looping-feedback
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through its core functionality. It processes external data from GitHub issues as direct input for an AI agent with modification capabilities.
- Ingestion points: The GitHub Actions workflow in
.github/workflows/feedback.ymlinterpolatesgithub.event.issue.titleandgithub.event.issue.bodydirectly into the agent's prompt. - Boundary markers: The prompt lacks delimiters or explicit security instructions to isolate the untrusted issue content from the system instructions.
- Capability inventory: The agent is granted the
Bashtool in the workflow configuration, and the workflow environment hascontents: writepermissions, allowing the agent to execute shell commands and modify the source code. - Sanitization: No sanitization or validation is performed on the issue content before it is passed to the agent.
- [EXTERNAL_DOWNLOADS]: The workflow fetches and executes the
anthropics/claude-code-action@v1GitHub Action. This is an official component provided by a trusted organization.
Audit Metadata