factory-review

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references an installation command using npx that targets the author's own repository jwilger/agent-skills. This is a standard and expected deployment pattern within the ecosystem.
  • [COMMAND_EXECUTION]: Provides functionality to update the local .factory/config.yaml file. These operations are guarded by validation logic and require explicit human confirmation before any changes are persisted, preventing unauthorized or accidental modifications.
  • [DATA_EXFILTRATION]: Accesses build-related metadata and logs located in the .factory/ directory. These read operations are strictly local and necessary for generating the dashboard summaries; no network patterns or outbound data transfers were observed.
  • [PROMPT_INJECTION]: Processes data from build retrospectives and PR summaries. While these sources include human-generated text, the skill operates on internal project artifacts with no identified intent to bypass safety guidelines or override agent behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 09:54 PM