skills/spm1001/trousse/consomme/Gen Agent Trust Hub

consomme

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and analyze untrusted data from BigQuery tables. Ingestion points: Data enters the agent context through the execute_sql and search_catalog tools in SKILL.md. Boundary markers: The instructions do not specify the use of delimiters or clear markers to isolate data content from system instructions. Capability inventory: The skill has access to the Bash, Read, and Write tools as defined in the allowed-tools frontmatter. Sanitization: No explicit sanitization or filtering of SQL results is prescribed before the data is processed by the agent.
  • [EXTERNAL_DOWNLOADS]: The dashboard generation templates reference well-known external libraries for visualization. Evidence: references/dashboard-patterns.md includes script tags for Chart.js and a date-fns adapter hosted on the jsDelivr CDN.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 08:46 AM
Security Audit — agent-trust-hub — consomme