Output and state - AgentGG

The output directory

A scan writes every result under one output directory.

out/
├── summary.md            ← human report
├── findings/...          ← one .md per finding (GHSA-shaped)
└── state/                ← what `status` and `revalidate` read
    ├── scan.json         ← root path + timestamps
    ├── recon.json        ← the recon brief (phase 1)
    ├── plan.json         ← which agents queued / skipped + why (phase 2)
    ├── runs/<id>.json    ← one per scan / recon / revalidate / score / summary
    ├── agents/<slug>.json← per-agent resume sidecar
    └── files/<path>.json ← FileRecord per scanned source file

summary.md

The human report for the scan.

findings/

One markdown file per finding. See Resume and reports for the file shape.

state/

File	Purpose
`scan.json`	The scan root path and its timestamps.
`recon.json`	The recon brief from phase 1.
`plan.json`	Which agents the plan queues or skips, and why. Phase 2 writes it before any agent runs.
`runs/<id>.json`	One record for each scan, recon, revalidate, score, or summary run.
`agents/<slug>.json`	The resume record for one agent.
`files/<path>.json`	The record for one scanned source file.

Why the state directory matters

Resume, status, and revalidate all read the state/ directory. A re-run with the same -o skips files that did not change. A different -o starts a fresh scan. See Resume and reports for what makes a re-run reuse this state.