notion-spec-to-implementation

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for configuring the local agent environment using shell commands during the setup process described in SKILL.md. Evidence: codex mcp add notion --url https://mcp.notion.com/mcp, codex --enable rmcp_client, codex mcp login notion.
  • [PROMPT_INJECTION]: The skill operates on an indirect prompt injection surface by ingesting content from external Notion pages (Notion:notion-fetch) and using it to drive automated actions like creating implementation plans and tasks (Notion:notion-create-pages). Evidence: 1. Ingestion points: Notion:notion-fetch calls in SKILL.md and reference/spec-parsing.md. 2. Boundary markers: Absent; there are no instructions to use delimiters or specifically ignore instructions embedded in the specs. 3. Capability inventory: Page creation and update capabilities via Notion:notion-create-pages and Notion:notion-update-page. 4. Sanitization: Absent; content from specifications is interpolated into new Notion pages without filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 12:25 PM
Security Audit — agent-trust-hub — notion-spec-to-implementation