platform-apex-anonymous-run

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly requires echoing any developer-pasted Apex snippet back for confirmation and writing it verbatim to a persistent temp file (and not auto-deleting it), so if those snippets contain API keys/passwords or other secrets the LLM must output and persist them verbatim, creating a high exfiltration risk.

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

• Third-party content exposure detected (low risk: 0.10). In platform-apex-anonymous-run, runtime anonymous Apex execution ingests outsider-authored free text only when the developer explicitly provides a pasted snippet (Step 2 snippet mode → Step 4 Write to .sfdx/tmp/anon-<ts>.apex, then Step 5 runs sf apex run --file <tmp-path>), and there is no automatic ingestion from external feeds/queues.