pflow-task-next
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No security issues were identified. The skill performs local repository management tasks using
gitandmdtodo. It implements input validation for the agent-generated branch names usinggit check-ref-formatand uses a non-executing parser for environment configuration files.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted task titles and descriptions from an external tool, which represents a potential injection surface.\n - Ingestion points: Task data is fetched from the
mdtodoCLI tool intask-next-context.sh.\n - Boundary markers: The
SKILL.mddefines strict naming conventions (type/slug) to constrain agent interpretation of task data.\n - Capability inventory: The skill executes
git checkout -bthroughtask-next-branch.shto create branches.\n - Sanitization: All branch names are validated using
git check-ref-formatbefore execution intask-next-branch.shto prevent command injection or invalid branch names.
Audit Metadata