handoffplan

Pass

Audited by Gen Agent Trust Hub on Oct 31, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted conversation history and session data to generate technical plans and metadata. \n
  • Ingestion points: Reads conversation history and session files (via the referenced /handoff skill) in SKILL.md. \n
  • Boundary markers: No explicit delimiters or instructions to ignore instructions embedded in the session data are provided. \n
  • Capability inventory: File system write operations for plans/handoffs, git commit capabilities, and task management via the bd tool. \n
  • Sanitization: The skill lacks explicit sanitization or escaping of the ingested data before it is interpolated into shell commands (such as bd create or git commit) or markdown plan files. \n- [COMMAND_EXECUTION]: The skill utilizes local CLI tools to manage project state and session persistence. \n
  • Evidence: Executes git status, git diff, git add, and git commit to version control session output. It also interfaces with a task management tool named bd using commands like create, dep add, update, and remember to track development phases.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 31, 2026, 03:29 PM