audit-performance
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external, untrusted sources which could potentially contain malicious instructions.\n
- Ingestion points: Scraped markdown content and search results from the web via
firecrawl, as well as transaction names and issue logs fromsentry, are loaded into the agent's context in SKILL.md.\n - Boundary markers: The instructions lack specific delimiters or guardrails to prevent the agent from following commands that might be hidden within these external data sources.\n
- Capability inventory: The skill is designed to execute shell commands (
npm,npx) and database queries (EXPLAIN ANALYZE), creating a path from data ingestion to command execution.\n - Sanitization: No filtering or sanitization steps are defined for the data retrieved from external APIs before it is used to guide the optimization process.\n- [COMMAND_EXECUTION]: The skill instructs the agent to run build scripts, bundle analysis tools, and SQL query analysis commands as part of the performance audit process.\n- [EXTERNAL_DOWNLOADS]: The skill uses
npxto install and run thesource-map-explorerpackage from the public npm registry for bundle analysis.
Audit Metadata