skills/etalab-ia/skills/rag-index/Gen Agent Trust Hub

rag-index

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the @tobilu/qmd package globally from the NPM registry using bun install -g. This is a third-party dependency required for the skill's primary functionality.
  • [COMMAND_EXECUTION]: The skill extensively uses shell commands to interact with the installed tool, including qmd collection add to create indices, qmd context add for metadata, and qmd embed for vector generation.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted local markdown data, which creates a vulnerability surface for indirect prompt injection if the documents contain malicious instructions.
  • Ingestion points: Reads documents from directory paths provided in the path argument (e.g., ./docs, ~/Documents/api-docs) via qmd collection add.
  • Boundary markers: None identified; the skill processes raw markdown content as provided.
  • Capability inventory: Executes shell commands via qmd and performs file system lookups (ls -la, find).
  • Sanitization: Not present; the tool creates embeddings based on the raw text content of the documents.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 02:48 PM