dineway-enrich-places
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes a Node.js script
scripts/enrich_place_details.jsto perform restaurant data enrichment, search, and authentication flows. - [EXTERNAL_DOWNLOADS]: The script fetches data from the vendor's official API endpoint at
https://api.dineway.ai. - [CREDENTIALS_UNSAFE]: The skill stores sensitive authentication tokens (access and refresh tokens) in a local file
.dineway/forgeway.json. While it applies best practices by setting file permissions to0o600and creating a.gitignoreto prevent accidental exposure in version control, this represents a local persistence of credentials. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by ingesting and saving data from an external source.
- Ingestion points: Data is retrieved from the Dineway API and saved to the local workspace via
scripts/enrich_place_details.js. - Boundary markers: The skill does not implement delimiters or safety instructions when processing or saving external API content.
- Capability inventory: The script has file-system write access (
fs.writeFileSync) and command execution capabilities within the environment. - Sanitization: While the script sanitizes output filenames to prevent path traversal, the content of the API response is saved to the workspace without sanitization, allowing potential instructions embedded in the API data to persist in the user environment.
Audit Metadata