local-backlog-manager

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes local development utilities such as git, grep, and glob to manage and query the project repository. These operations are limited to the local filesystem and are consistent with the skill's purpose of project tracking.
  • [PROMPT_INJECTION]: The skill processes user-provided specifications to generate structured markdown work items. While this constitutes an indirect prompt injection surface, the risk is mitigated by a mandatory human-in-the-loop preview step (Step 3) where the user reviews the generated IDs, estimates, and metadata before any files are created or modified.
  • Ingestion points: documentation/specs/{feature}/tasks.md (SKILL.md)
  • Boundary markers: Absent in the parsing logic, but human review is required before publishing.
  • Capability inventory: File write and edit access to markdown files within the documentation/tracker/ and documentation/specs/ directories.
  • Sanitization: Not explicitly defined for the input data parsing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 05:15 AM
Security Audit — agent-trust-hub — local-backlog-manager