context-xray
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The setup instructions utilize
npxto download the@agent-native/corepackage from the public npm registry. \n- [REMOTE_CODE_EXECUTION]: The skill executes the downloaded@agent-native/corepackage vianpxto install local command and skill artifacts. \n- [COMMAND_EXECUTION]: Provides instructions to run thecontext-xrayCLI tool locally to analyze thread data and open generated HTML reports in a web browser. \n- [PROMPT_INJECTION]: The skill processes local conversation transcripts (untrusted data) to generate summaries and optimizations, creating an indirect prompt injection surface. \n - Ingestion points: Reads local transcript files via the
context-xraytool (SKILL.md). \n - Boundary markers: No explicit delimiters or warnings are provided to the agent to ignore instructions embedded within the transcripts. \n
- Capability inventory: The skill executes local CLI commands and opens files in the web browser. \n
- Sanitization: There is no evidence of input validation or escaping for the transcript content before it is processed or summarized.
Audit Metadata