What problem does it solve? Bug bounty researchers need a systematic way to find information disclosure vulnerabilities—leaked tokens, PII, hidden JSON fields, and exposed CI logs—without over-claiming or violating program rules. This Skill provides a structured hunting method with verified negative controls so findings are real, in-scope, and reportable. ## Core Features & Use Cases - Serialization Drift Testing: Diff HTML, JSON, GraphQL, export, and print representations of the same object to find fields the requesting identity should not receive. - Public Artifact & Log Sweep: Inspect CI logs, bot-created repositories, source maps, and build artifacts for active secrets, then confirm with read-only identity endpoints only. - Cross-Origin & Parser Isolation Tests: Detect authenticated data in dynamic JavaScript (e.g., service workers) and memory leakage through file preview/conversion pipelines using canary uploads. - Negative-Control Taxonomy: Apply 12 stop-condition categories grounded in real HackerOne report outcomes to avoid filing duplicates, expected behavior, or below-threshold findings. - Use Case: While testing a target's report endpoint, request its .json variant with a low-privilege token, diff field names against the authorized UI, and discover internal reporter attributes—then redact and report with evidence. ## Quick Start Ask the agent to test a target object's JSON and GraphQL representations for fields beyond what a low-privilege account should see, using the info-disclosure hunting method.
