exa-upgrade-migration

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill analyzes and modifies local project files to perform SDK upgrades, which exposes a surface for indirect prompt injection from untrusted code within the repository.\n
  • Ingestion points: Reads project source code to detect breaking changes as described in Step 4.\n
  • Boundary markers: There are no explicit instructions or delimiters used to separate user code from agent instructions during the migration analysis.\n
  • Capability inventory: Access to 'Edit' and 'Write' for file modification, and 'Bash' for package installation.\n
  • Sanitization: No sanitization or validation of the ingested code content is performed.\n- [EXTERNAL_DOWNLOADS]: Fetches the '@exa/sdk' package from the npm registry and references official Exa documentation and GitHub repositories.\n- [COMMAND_EXECUTION]: Executes npm commands (list, view, install, test) and git commands to manage project dependencies and version control branches.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 12:20 PM