context7-cli
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the
ctx7CLI globally usingnpm install -g ctx7@latestor to run it directly vianpx. This is the standard delivery method for the tool that powers the skill's functionality. - [COMMAND_EXECUTION]: All features, including library searching, ID resolution, and documentation fetching, are executed via shell commands using the
ctx7CLI. It also manages agent configuration and environment variables (CONTEXT7_API_KEY) through the command line. - [INDIRECT_PROMPT_INJECTION]: The skill has an inherent attack surface for indirect prompt injection because it fetches external content (documentation and snippets) based on user-provided queries. The author addresses this risk with explicit safety warnings.
- Ingestion points: Data is ingested from the Context7 documentation service via the
ctx7 docs,ctx7 library, andctx7 searchcommands. - Boundary markers: No specific technical delimiters are enforced in the command templates, but the instructions provide guidance on result selection and topic specificity.
- Capability inventory: The agent can execute shell commands, access the network via the CLI, and modify local configuration files during the setup process.
- Sanitization: The skill includes clear instructions for the user and agent to omit sensitive or proprietary information from search queries to prevent accidental data exposure.
Audit Metadata