notion-spec-to-implementation

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process potentially untrusted data from external Notion specification pages to generate project artifacts.\n
  • Ingestion points: Data enters the agent context through the Notion:notion-fetch tool as described in SKILL.md (Step 1) and reference/spec-parsing.md.\n
  • Boundary markers: The templates provided in reference/standard-implementation-plan.md and reference/task-creation-template.md do not include explicit delimiters or system-level instructions to ignore embedded agent commands within the parsed specification text.\n
  • Capability inventory: The skill possesses significant write capabilities within the user's Notion environment via Notion:notion-create-pages and Notion:notion-update-page tools, which are used to generate plans and tasks based on parsed requirements as defined in SKILL.md and reference/task-creation.md.\n
  • Sanitization: There is no evidence of content sanitization or validation logic to filter out steering instructions or non-requirement content from the source specifications before they are used to generate plan descriptions or task acceptance criteria.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 09:56 AM
Security Audit — agent-trust-hub — notion-spec-to-implementation