skills/broomva/skills/handoff/Gen Agent Trust Hub

handoff

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates the transfer of session state by drafting markdown documents that include repository metadata and executable command blocks for subsequent agents. While this creates a surface for indirect prompt injection—where potentially untrusted data from the current session (such as commit messages or PR descriptions) could be carried over into commands suggested to a future session—this behavior is the primary intended function of the skill and does not involve malicious logic.
  • Ingestion points: Current agent context, git repository metadata (status, logs), and PR information as defined in SKILL.md and references/handoff-template.md.
  • Boundary markers: The skill uses markdown headers and structured sections to organize state but does not provide explicit instructions to sanitize or ignore instructions embedded in the source metadata.
  • Capability inventory: The skill is a documentation generator and does not execute commands directly; however, it documents state and provides commands intended for subsequent session execution.
  • Sanitization: No explicit sanitization or filtering of session metadata is performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 03:33 PM
Security Audit — agent-trust-hub — handoff