sf-debug

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and parses debug logs from target Salesforce orgs (e.g., via "sf apex get log" / "sf apex tail log" as described in SKILL.md and cli-commands.md) and a registered hook (hooks/scripts/parse-debug-log.py, wired in .claude/hooks.yaml) ingests that user-generated/untrusted log content and produces agent-facing analysis and agentic fix recommendations, so third-party content can materially influence tool use and next actions.