find-skills
Warn
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates downloading content from external sources via the
npx playbooks add skill <source>command. These sources are dynamic and not restricted to a trusted list. - [COMMAND_EXECUTION]: The skill relies on executing shell commands using
npx, which downloads and runs Node.js packages from the npm registry. - [REMOTE_CODE_EXECUTION]: The primary purpose of the skill is to install new 'agent skills,' which are executable extensions for the AI agent. This introduces a risk of remote code execution if a malicious source is provided.
- [INDIRECT_PROMPT_INJECTION]: The skill performs lexical and semantic searches (
--semantic) on external skill repositories. Maliciously crafted skill names or descriptions in these repositories could influence the agent's behavior during the discovery phase. - [AUTONOMY_CONCERNS]: Although the instructions mention asking the user for confirmation, the use of the
-yflag in the installation examples suggests a pattern for bypassing interactive prompts, which could be exploited to install skills without sufficient review.
Audit Metadata