context-mode
Processes large outputs in sandboxed subprocesses, keeping only summaries in context for ~98% context savings.
🛡️ AgentReady threat assessment
MAESTRO 7-layer threat model + OWASP AIVSS risk score for context-mode, derived from its capabilities.
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology — every score is re-derived by the same automated method as an agent's public evidence changes.
Overview
context-mode runs large tool outputs in sandboxed subprocesses and keeps only summaries in the Claude Code context window, reporting up to 98% context savings across 21 benchmarked scenarios. Its subprocess-sandboxing design has direct execution/security surface.
Key features and capabilities
- Sandboxed subprocess execution
- Summary-only context retention
- ~98% context savings benchmarked
Use cases
- Handle huge command outputs without blowing context
- Keep long sessions within the context budget