find-docs
Warn
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the installation of an external package from the NPM registry using
npm install -g ctx7@latest. - [REMOTE_CODE_EXECUTION]: The skill executes remote code via
npx ctx7@latest, which fetches and runs the package from the registry at runtime. - [COMMAND_EXECUTION]: The skill relies on shell command execution for its core functionality, including
ctx7 library,ctx7 docs, andctx7 login. - [PROMPT_INJECTION]: The skill ingests untrusted content from the
ctx7CLI output, creating a surface for indirect prompt injection. - Ingestion points: Library metadata and documentation snippets retrieved via
ctx7 libraryandctx7 docs(SKILL.md). - Boundary markers: Absent. The skill provides no instructions to separate external data from system prompts or to ignore embedded instructions.
- Capability inventory: Shell command execution through the
ctx7tool. - Sanitization: Absent. The retrieved content is processed directly without validation or escaping.
Audit Metadata