autonomous-agent-harness
Pass
Audited by Gen Agent Trust Hub on Jul 4, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill configures the agent to use official MCP servers from a trusted organization, including
@anthropic/memory-mcp-server,@anthropic/scheduled-tasks-mcp-server, and@anthropic/computer-use-mcp-server. These are legitimate platform extensions. - [COMMAND_EXECUTION]: Provides instructions for using programmatic mode (
claude -p) and scheduled task creation via thescheduled-tasksMCP tool. These commands are used as intended for autonomous workflow management and do not involve unauthorized shell execution or privilege escalation. - [DATA_EXFILTRATION]: Documents the use of the Anthropic Dispatch API for remote agent triggers. The use of
$ANTHROPIC_API_KEYfollows security best practices for environment variable usage rather than hardcoding credentials. - [PROMPT_INJECTION]: The skill defines agent personas and workflows but does not attempt to override system safety guidelines or bypass AI constraints.
Audit Metadata