consensus-summary-html
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the agent to execute a local Python validation script (
scripts/validate_consensus_summary.py) to verify the integrity of the generated HTML output. This script is provided within the skill's directory. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data (model consensus reports, state files, and round digests). This creates an attack surface where malicious content within the processed data could influence the summary or be rendered into the resulting HTML brief.
- Ingestion points: Reads content from
report_path(Markdown),state_path(JSON), and round output files. - Boundary markers: The skill mandates a specific HTML structure with defined IDs and data roles to compartmentalize the output.
- Capability inventory: Performs file writes to create the HTML summary and executes a local Python script for validation.
- Sanitization: Instructions specify that output must follow shared HTML page conventions which include escaping content.
Audit Metadata