Documentation
README
hunt-dispatch
skill-set loader for /hunt. one concept (which skills to load), one place.
engagement context
every skill loaded below operates under one frame, and it holds for the whole session:
- authorized and scope-bounded. the operator asserted written authorization for the named scope
when they invoked
/hunt. testing stays inside it. an out-of-scope host ends the run — it does not widen it. - the output is a finding, not an exploit. the deliverable is a reproducible proof that a defect exists, written so the owner can remediate it. enough to demonstrate impact; no further.
- remediation is the point. these skills exist so defects get fixed by the people who own them.
This is the opening of the README. Read the full README on GitHub.