skills-discovery

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • REMOTE_CODE_EXECUTION (MEDIUM): The skill uses npx skills-installer to execute external code from an untrusted package registry. This represents a supply chain risk where the installer or the npm registry could serve malicious content.
  • EXTERNAL_DOWNLOADS (MEDIUM): The skill performs HTTP requests to https://claude-plugins.dev/api/skills, an unverified third-party domain not included in the trusted sources list.
  • DYNAMIC_EXECUTION (MEDIUM): The skill's primary purpose is to dynamically install and load new scripts based on external data, which creates a risk of loading and executing malicious capabilities into the agent's runtime environment.
  • PROMPT_INJECTION (LOW): The skill is vulnerable to indirect prompt injection because it ingests and processes unverified skill names and descriptions from an external API. Evidence Chain: 1. Ingestion points: curl response from the registry API. 2. Boundary markers: Absent. 3. Capability inventory: Command execution (npx, curl). 4. Sanitization: No evidence of validation or escaping for the data returned by the API.
  • COMMAND_EXECUTION (LOW): The skill relies on shell command execution (curl, npx) to perform its core management and discovery tasks.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 06:44 PM