skill-optimizer

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly requires extracting and quoting "user messages (full text)" as evidence and reporting which user messages should have triggered a skill, which forces the LLM to handle and potentially output verbatim sensitive tokens (API keys, passwords, cookies) present in transcripts.

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

• Third-party content exposure detected (high risk: 0.85). Outsider-authored free text is ingested because the workflow scans runtime session JSONL transcripts from ~/.claude/projects/**/*.jsonl and ~/.codex/sessions/**/*.jsonl, extracting “User messages (full text)” and then uses that text to compute dimensions like trigger rate and undertrigger detection, which feeds into the agent’s LLM context for the generated diagnostic report.