agile-review
Pass
Audited by Gen Agent Trust Hub on May 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The instructions and templates provided are focused on standard sprint review tasks. There is no evidence of obfuscation, remote code execution, or unauthorized network activity.\n- [PROMPT_INJECTION]: The skill processes external data such as sprint issues and status reports, creating a surface for indirect prompt injection. However, since the skill's capabilities are limited to generating markdown documentation and do not include the execution of dangerous commands or network operations, the risk is minimal.\n
- Ingestion points: Reads sprint issues, status closure reports, and checkpoints as defined in SKILL.md.\n
- Boundary markers: No explicit markers or instructions are provided to the agent to distinguish between its instructions and the data being processed.\n
- Capability inventory: The skill only performs text summarization and markdown formatting; it has no access to subprocesses, sensitive file systems, or external networks.\n
- Sanitization: No sanitization or validation of the input data is specified in the skill's logic.
Audit Metadata