skills/boshu2/agentops/rch/Gen Agent Trust Hub

rch

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local system commands (rch, ssh, rsync) and modifies local configuration files, including the agent's settings file (~/.claude/settings.json) to install hooks for command interception and routing.
  • [REMOTE_CODE_EXECUTION]: The skill's primary function is to execute arbitrary compilation commands on remote worker machines using the rch exec utility.
  • [EXTERNAL_DOWNLOADS]: The skill manages remote toolchains and can trigger the download and installation of packages (e.g., Rustup, Cargo) on remote workers via the rch workers setup and sync-toolchain commands.
  • [CREDENTIALS_UNSAFE]: The skill identifies and accesses local SSH private keys (e.g., ~/.ssh/id_ed25519) for worker authentication as defined in its worker configuration files.
  • [DATA_EXFILTRATION]: The skill accesses sensitive configuration and identity files to perform its duties; however, these operations are directed towards configured remote workers for legitimate compilation tasks, and no evidence of data exfiltration to unauthorized domains was detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 04:20 PM
Security Audit — agent-trust-hub — rch