fr-triagelisted

# fr-triage **Announce at start:** "I'm using fr-triage to triage <repo or org>." You can already read an issue, check it against the code and rank it. A triage done in chat loses the rest: the result dies with the session, syncs are invisible, a refresh redoes it all. `fr triage` keeps your ranking in a file. ## State: two files, two owners Everything lives in `$HOME/.cache/fr/triage/<scope>/` (`owner--repo` for `--repo OWNER/REPO`, `owner` for `--org OWNER`, lowercased; `--dir D` overrides). Pass the same `--repo`/`--org`/`--dir` to every command. | File | Written by | Holds | |---|---|---| | `facts.json` | `fr triage collect` | what the forge says: open issues, labels, linked PRs (no stages) | | `judgements.yaml` | **you**, plus the `batch` verbs for `batches:` | tiers, per-issue rankings, patterns, batches | | `triage.html` | `fr triage render` | the board, built from both | Stages (`backlog`, `blocked`, `in-progress`, `pr-draft`, `pr-ready`, `merged`, `closed`) are derived by `check` and `render`, never stored. Never set one, and never write facts yourself. ## The loop (a re-run of it is the sync) 1. **Collect.** `fr triage collect --repo OWNER/REPO` (or `--org OWNER`). If it warns that the **PR list** hit its limit, re-run once with `--pr-limit 1000`; if it still warns, go on and say so in your report. An issue- or repo-list warning has no flag: go on, and say rows may be missing. 2. **Check.** `fr triage check --repo OWNER/REPO [--json]` prints six sets a