cargo-triage

What problem does it solve? Incident analyses and proposals arrive in a shared intake from multiple repositories, and without a defined triage process their state is lost between sessions: records stay unread, fixes go unmarked, and releases are never linked back to the reports that triggered them. ## Core Features & Use Cases - State-driven triage flow: Moves cargo records through new, in-work, done, released, and quarantine states using read and mark commands, with batch operations and refusal handling. - Fix and release tracking: Requires every move to done to carry a fix description and every release to name a version, so each record answers what fixed it and where to find the fix. - Incident analysis discipline: Defines how incident analyses are written, named, sent, and removed from local disk so the intake holds the single authoritative record. - Use Case: After merging a fix for a reported incident, an agent reads the intake filtered by new records, creates a task, and marks the record done with the fix description in the same turn, then the publisher marks it released with the version. ## Quick Start Read the new cargo records from the intake and triage each one by creating a task and marking its state according to the cargo-triage rule.