autocontext-consumer

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the autoctx CLI tool and standard utilities like cat. These commands are used for listing, showing, exporting, and importing knowledge data.
  • [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by instructing the agent to read and process content from external files such as playbook.md and lessons.json which may contain untrusted instructions.
  • Ingestion points: Files located within the scenario directory (e.g., playbook.md, lessons.json, hints.md).
  • Boundary markers: None identified in the skill instructions to distinguish between data and instructions.
  • Capability inventory: The skill allows the agent to execute shell commands (cat, autoctx) and perform filesystem operations.
  • Sanitization: No sanitization or validation of the content of the knowledge files is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 09:00 AM
Security Audit — agent-trust-hub — autocontext-consumer