synthesis-checkpoint

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the date command to establish an authoritative time anchor and git log to verify the chronological history of project changes. It also invokes the synthesis-agent-conformance CLI tool for system health checks.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes various project files including CONTEXT.md, REFERENCE.md, and session logs, which are external data sources that could theoretically contain instructions intended to influence the agent.
  • Ingestion points: Files located at ~/.synthesis/coordination/active-sessions.md, CONTEXT.md, and sessions/YYYY-MM.md are read into the agent's context during the checkpoint process.
  • Boundary markers: The protocol does not specify the use of specific boundary markers or delimiters when reading these files.
  • Capability inventory: The skill has the ability to execute shell commands (date, git), write to project files (CONTEXT.md), and run a local utility tool (synthesis-agent-conformance).
  • Sanitization: There is no explicit mention of sanitizing or filtering the content of these files before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 10:03 AM