jaspar-database

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow fetches JASPAR REST API JSON from https://jaspar.elixir.no/api/v1/matrix/ and .../matrix/{id}/, which is public web content authored/maintained by an external third party, and that JSON (PFM fields like pfm arrays and metadata) is ingested into the agent’s LLM context as readable text.