The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could contain malicious instructions. • Ingestion points: Reads project_brief and agent_report as inputs in SKILL.md. • Boundary markers: Absent; no delimiters or instructions to ignore embedded commands are present. • Capability inventory: Orchestrates phase transitions, gate validations, and task distribution across multiple system agents (e.g., requirements, developer, security agents). • Sanitization: Absent; no logic is present to validate or filter the content of incoming reports or briefs.
[SAFE]: No hardcoded credentials or exposure of sensitive files (e.g., SSH keys, environment variables) were found.
[SAFE]: No remote code execution patterns, unauthorized command execution, or obfuscated content were detected.

workflow-management