Skills
Navigate
skills/getpaseo/paseo/paseo-plugin/Gen Agent Trust Hub

paseo-plugin

Pass

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes CLI commands including paseo plugin init, npm install, paseo reload, and rg to manage the plugin lifecycle and audit code.
  • [EXTERNAL_DOWNLOADS]: The skill fetches documentation from paseo.sh and supports plugin installation from npm and github registries.
  • [DYNAMIC_EXECUTION]: The skill facilitates the creation and execution of plugin code, which runs as unsandboxed daemon subprocesses.
  • [PRIVILEGE_ESCALATION]: The skill provides instructions to enable the pluginsEnabled setting in the daemon configuration, allowing unsandboxed code execution. This is accompanied by a mandatory user security warning and consent check.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data from user requests and remote documentation. Ingestion points include $ARGUMENTS and https://paseo.sh/llms.txt. Boundary markers are used to label user input. The skill possesses capabilities for command execution and file manipulation. No explicit sanitization is performed beyond relying on user inspection of generated code.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 1, 2026, 09:22 AM
Security Audit — agent-trust-hub — paseo-plugin