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

paseo-plugin

Pass

Audited by Gen Agent Trust Hub on Nov 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is instructed to fetch and follow external documentation from https://paseo.sh/llms.txt and related Markdown pages at runtime to catch contract changes.
  • Ingestion points: External URLs including https://paseo.sh/llms.txt, https://paseo.sh/docs/plugins.md, and https://paseo.sh/docs/plugins/reference.md (specified in SKILL.md).
  • Boundary markers: Absent. The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing these external files.
  • Capability inventory: The skill possesses the ability to write files to the filesystem, execute shell commands (via npm and paseo CLI), and install plugins from remote sources.
  • Sanitization: Absent. There is no mention of validating or sanitizing the content of the fetched documentation before the agent adopts its instructions.
  • [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute various shell commands on the host system.
  • Evidence: Use of npm install, paseo plugin init, paseo plugin install, paseo plugin reload, and rg (ripgrep) for auditing code.
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches content from external sources associated with the vendor.
  • Evidence: Fetches documentation and manifests from paseo.sh and provides instructions to clone repositories from github.com/getpaseo.
  • [REMOTE_CODE_EXECUTION]: The skill facilitates the installation and execution of code from remote registries.
  • Evidence: Instructions to run paseo plugin install github:owner/repository and paseo plugin install npm:@acme/paseo-review, which downloads and runs unsandboxed code as a daemon subprocess.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 19, 2026, 02:46 AM