osint-autopilot

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to fetch exposed secrets (e.g., from JS/vars, S3 checks, JS harvest) and produce a consolidated workbook with a "Secrets" tab, which requires including secret values verbatim in output/files.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This repo is an explicit offensive OSINT/autopilot tool that intentionally harvests secrets from public JS, performs active port/content scans, checks for S3 takeovers and breach credentials, and synthesizes a ranked "look-here-next" queue for exploitation — functionality that clearly enables deliberate misuse beyond benign observation.

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

• Third-party content exposure detected (medium risk: 0.30). Outsider-authored free text is ingested at runtime via the Stage 4 fan-out workflow reading bucket file lists produced from the recon pipeline’s HTTP/JS harvesting (e.g., ~/Research/engagements/<domain>/buckets/bucket-* and then synthesizing with JSON.stringify(all)), even though it is “derived evidence” rather than first-party docs.