research-deprecation-usage
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Node.js script located at
scripts/findDeprecations.mjsusingyarn node. This script is used to extract deprecation metadata from the local monorepo for auditing purposes. - [INDIRECT_PROMPT_INJECTION]: The skill ingests external code snippets from customer repositories via Sourcegraph searches to identify API usage. This ingestion of untrusted data creates a potential surface for indirect prompt injection.
- Ingestion points: External source code hits returned by the Sourcegraph MCP keyword search (Step 3).
- Boundary markers: The instructions do not specify any boundary markers or "ignore embedded instructions" warnings for the external code being processed.
- Capability inventory: The skill has access to local command execution (
yarn node) and external search tools via the Sourcegraph MCP server. - Sanitization: There is no evidence of sanitization, filtering, or validation performed on the code content retrieved from Sourcegraph before it is processed by the agent.
Audit Metadata