context7-cli
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill operates by instructing the agent to execute the
ctx7CLI tool to search for documentation, install skills, and configure environment settings. These commands are executed within the local development environment. - [EXTERNAL_DOWNLOADS]: The skill utilizes the npm registry to install the
ctx7package globally or run it vianpx. Asctx7is a resource provided by the vendor Upstash, this aligns with standard deployment practices for developer tools. - [INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves fetching external documentation, which introduces a potential surface for indirect prompt injection.
- Ingestion points: The agent ingests external documentation and code snippets via the
ctx7 libraryandctx7 docscommands. - Boundary markers: Instructions are provided to limit documentation calls, though no specific structural delimiters for the tool's output are defined in the skill content.
- Capability inventory: The skill has the ability to execute shell commands (
ctx7) and modify agent configuration files (such as.mcp.json) via thectx7 setupcommand. - Sanitization: The skill instructions explicitly warn the agent not to include sensitive credentials or proprietary code in search queries, providing a layer of operational safety.
Audit Metadata