kw-find-docs
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends installing or running the
ctx7CLI tool vianpm install -g ctx7@latestornpx ctx7@latest. This involves downloading and executing code from the npm registry. - [COMMAND_EXECUTION]: The skill utilizes shell commands (
ctx7 libraryandctx7 docs) to search for and retrieve technical documentation. Usage is restricted to three attempts per query to manage resource consumption. - [DATA_EXFILTRATION]: The skill supports authentication via the
CONTEXT7_API_KEYenvironment variable and thectx7 logincommand. It includes explicit instructions for the agent to avoid including sensitive data, such as credentials or proprietary source code, in search queries. - [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection because it processes content retrieved from external sources.
- Ingestion points: Technical documentation and code examples retrieved via the
ctx7 docscommand in SKILL.md. - Boundary markers: Absent; there are no specific markers used to delineate external content from the agent's instructions.
- Capability inventory: Execution of shell commands via the
ctx7CLI tool. - Sanitization: Absent; the skill does not specify mechanisms for filtering or sanitizing the retrieved documentation before processing.
Audit Metadata