blueprint-curate-docs

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes bash commands for project introspection and metadata management. It uses find and ls to locate documentation files in the docs/blueprint/ directory, grep to check dependency versions and search the src/ directory for code patterns, and jq to update execution statistics in docs/blueprint/manifest.json.\n- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted content from the web. Ingestion points: Data is gathered via WebSearch and WebFetch in Step 2 to identify library 'gotchas' and documentation. Capability inventory: The skill is authorized to Write files and execute Bash commands. Boundary markers: No specific delimiters are defined in the instructions to isolate gathered external content from the rest of the documentation. Sanitization: The skill does not describe an explicit sanitization process for external documentation before it is integrated into the output files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 08:50 PM