log-decision

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill reads from a local configuration file provided by the vendor to identify the target workspace.
  • Ingestion points: .claude/jupi.local.json (SKILL.md).
  • Context: The skill retrieves a workspace identifier (groupSlug) from a local file. This is a standard practice for configuring local environment tools, and the skill includes instructions to prevent the file from being committed to version control.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user-provided text which is then recorded in an external system, creating a data ingestion surface.
  • Ingestion points: User-provided text for decision titles, options, and rationales (SKILL.md).
  • Boundary markers: None identified.
  • Capability inventory: Tools for creating and finalizing decision records (create-decision-tool, add-decision-options-tool, finalize-decision-tool).
  • Sanitization: The instructions recommend formatting descriptions as HTML for display, but no security-specific input filtering is performed. This behavior is consistent with the skill's primary function of recording user input.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 04:48 PM
Security Audit — agent-trust-hub — log-decision