fix-all-issues
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions direct the agent to identify and execute arbitrary commands found in repository files such as
package.json,Makefile,justfile, andAGENTS.md. Executing untrusted scripts from these locations can lead to unintended command execution if the repository contains malicious configuration. - [PROMPT_INJECTION]: The skill analyzes external data (code diffs and repository documentation) and incorporates it into the agent's context. There are no boundary markers or instructions to treat this content as untrusted, making the skill susceptible to indirect prompt injection where malicious instructions hidden in code comments or PR descriptions could influence agent behavior.
- Ingestion points: PR branch code,
package.json,Makefile,AGENTS.md, and CI configuration files. - Boundary markers: Absent. The instructions do not specify how to distinguish between repository content and orchestrator instructions.
- Capability inventory: Shell execution of discovered scripts (lint, test, build),
gitoperations includingpush --force-with-lease, and GitHub CLI (gh) operations. - Sanitization: Absent. The agent is encouraged to follow the instructions and logic found within the repository content directly.
Audit Metadata