parallel-data-enrichment
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill orchestrates the use of
parallel-clito perform data enrichment tasks, including suggesting columns, running enrichment jobs, and polling for results. These operations are standard interactions with the vendor's tools. - [DATA_EXFILTRATION]: The skill transmits user-provided datasets (via inline JSON or CSV files) to the vendor's backend for processing. This data movement is the primary, documented function of the enrichment service.
- [EXTERNAL_DOWNLOADS]: The skill provides automated setup via
/parallel:parallel-cli-setupand suggests updates for theparallel-web-toolspackage. These resources are authored by the vendor and are necessary for the skill's operation. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data provided by the user.
- Ingestion points: Data enters the skill via the
$ARGUMENTSplaceholder, user-supplied CSV files, and inline JSON objects. - Boundary markers: There are no explicit instructions or delimiters used to separate untrusted data from instructions within the tool calls.
- Capability inventory: The skill uses Bash to execute command-line tools and accesses the file system to save and read JSON results.
- Sanitization: No specific sanitization or filtering is described for the data being enriched, though this is expected behavior for a data processing utility.
Audit Metadata