The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data from various sources and processing it without adequate security boundaries.
Ingestion points: Reads project documentation from production/milestones/, production/sprints/, and production/risk-register/, and performs a Grep scan for TODO, FIXME, and HACK markers across the entire codebase.
Boundary markers: The skill lacks explicit boundary markers or instructions to the agent to disregard potential instructions embedded within the ingested text.
Capability inventory: The skill has access to the Write tool, creating a potential risk if malicious instructions found in code comments or reports were to be executed by the agent.
Sanitization: No validation or escaping is performed on the ingested content before it is interpolated into the generated markdown report.
[SAFE]: The skill's core logic is consistent with its stated purpose of milestone review and project status reporting.
[SAFE]: No direct prompt injection, obfuscation, or hardcoded credentials were found in the skill's instructions or metadata.

milestone-review