matrix-handoff

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the agent to capture and then re-run a command containing the exact approval token printed by the preview, which requires reproducing a sensitive token verbatim and creates an exfiltration risk.

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

• Third-party content exposure detected (medium risk: 0.30). scripts/matrix-handoff.mjs ingests outsider-authored free text indirectly by reading and packaging the optional user-supplied continuation brief file and (when enabled/selected) copying a raw Codex/Claude JSONL transcript from the user’s local filesystem into the handoff bundle for Matrix.