@r1z4x/tezgah 1.3.0 → 1.5.0

package/CHANGELOG.md

@@ -6,6 +6,106 @@ adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ## [Unreleased]
 
+## [1.5.0] - 2026-10-09
+
+### Changed
+
+- **A plan gets one review, at most two rounds.** The review reads the plan's
+  whole diff once, in its verification phase. Round one spawns its reviewers in
+  one message, one per dimension, or one reviewer for a diff under 300 lines.
+  Only a confirmed critical or major finding opens round two. Round two reads
+  the fix's delta. A minor finding goes on the fix-later list. `tezgah-task review`
+  refuses a third round. What is left after round two goes to the user. The
+  reviewer agent, the contract and plan-sync now state this one rule. Before,
+  the reviewer looped until a round confirmed nothing.
+- **One Acceptance rule.** A plan enters a writing phase only when each
+  Acceptance item names its command or says `unverifiable` and why. Before, one
+  command-bearing item let the move through. `render_table --acceptance
+  --strict` then refused the same plan. Both now read one predicate,
+  `tezgah_task.unproven`, and the refusal names each item's line.
+- **Typed decisions take any Jev carrier.** Taste's `learn`, `label`,
+  `calibrate`, `gate` and `measure --typed` require the Jev model, not
+  TypeSafe's address. So do taste's background start and replay's
+  `--label-model`. OpenRouter System One, Cloudflare Workers AI or a
+  compatible endpoint counts the same. Old `typesafe` rows keep counting.
+  `learn`'s report key `typesafe` is now `jev`. Replay's `--provider typesafe`
+  is now `--provider jev`, and the old name still works. The triage's no-judge
+  message names every carrier, the selection knob and the session CLIs.
+- **The fan-out rule arms on the lists the owner writes.** Items written
+  `1 - ...`, `1- `, `1:`, `a)` or `+` now arm it. An inline `1. ... 2. ...`
+  arms it, and so does "subagentlerle". Three files named as one list arm it
+  too, and so do "First, ... Second, ..." and three imperative clauses. A later
+  list in the same session gets a line that names the action: count the items,
+  one subagent per item in one message. The rule covers every phase: a plan's
+  tasks, its acceptance commands and a survey's sources fan out too. The clause
+  that let a list of small items run one by one is gone.
+
+### Added
+
+- **Plan order and progress in `tezgah-task status`.** A plan can name the
+  plans it waits on with `after:`. `status` prints each open plan's ticked
+  Acceptance boxes and what it still waits on. A `parallel:` line names the
+  ready plans whose `allowed_paths:` share no prefix. Those can run at once.
+  An open plan whose `status:` is not open or blocked is a `FAIL` line, and so
+  is an `after:` id no plan carries. `--json` adds `open` and `parallel`.
+- **The judge asks opencode, Cursor and Codex sessions too.** The judgement
+  seam's session CLI was omp or claude only. It now also runs `opencode run`,
+  `cursor-agent -p` and `codex exec` headless when their host's marker is set
+  (`OPENCODE`, `CURSOR_AGENT`/`CURSOR_VERSION`, `CODEX_THREAD_ID`).
+  `TEZGAH_JUDGE_CLI`, else `judge_cli` in `config.json`, picks one of the five
+  by name. A picked CLI that is not installed means no session CLI.
+- **Jev through any carrier.** The judge reaches Jev through TypeSafe,
+  OpenRouter, Cloudflare Workers AI or a System One-compatible URL
+  (`JEV_API_BASE_URL`, `JEV_API_KEY`).
+  `TEZGAH_JEV_PROVIDER`, `JEV_PROVIDER` or config.json `jev` picks the carrier,
+  and `tezgah-setup --jev <value>` writes the config key. `auto` asks every
+  carrier that resolves. A named carrier without its credential gets no
+  request, and no other carrier stands in. `tezgah-status --judge` prints the
+  choice and its carriers.
+- **The omp login is a TypeSafe key.** With no `TYPESAFE_API_KEY` and no key
+  file, the seam reads the active omp login, read-only.
+- **Clef and Clef-flash as typed judges.** Cloudflare's own decision models
+  answer the System One API on Workers AI with the Cloudflare token and account
+  id. `--jev clef` or `--jev clef-flash` picks one, and `auto` asks them after
+  the Jev carriers. A typed decision counts them like Jev. Cost rows now price
+  the carrier that answered: $0.24 per 1M input for Clef, $0.09 for Clef-flash.
+
+### Fixed
+
+- **A spent key stops costing a request.** OpenRouter answers 403 to a key
+  past its limit. A 403 now marks a provider down for five minutes, like a 401.
+
+## [1.4.0] - 2026-10-08
+
+### Changed
+
+- **Taste state lives in SQLite.** The repository's signals, decisions,
+  defects, labels, injections, gate state, learnings and write-note history
+  move to `.tezgah/taste/taste.db`. User-scope learnings move to
+  `~/.config/tezgah/taste/taste.db`. Opening a store imports the old JSON and
+  JSONL files and renames each to `<name>.imported`. It also takes rows an
+  older install appends to an old file later. `taste.md` stays a text file.
+- **The evidence ledger lives in SQLite.** Session rows move from
+  `~/.cache/tezgah/evidence/<session>.jsonl` to the `evidence` table of
+  `~/.cache/tezgah/tezgah.db`. Each row is one INSERT, so no reader meets a
+  torn line. A read or write of a session imports its old JSONL file first.
+  Session start imports the rest in the background once a day, and
+  `tezgah-doctor --import-evidence` imports them now. opencode's plugin uses
+  the same database through node:sqlite. On node before 22.5 it calls the
+  store's `evidence` CLI. The old files stay in place: an opencode process
+  started before the upgrade may still append to one. `tezgah-doctor --clean`
+  deletes a session idle past the retention window.
+- **The small cache stores live in the same database.** They move from files
+  under `~/.cache/tezgah/` to tables of `~/.cache/tezgah/tezgah.db`. These are
+  the turn stamps, switch baselines, used-tool marks and status marks. The
+  nudge marks, the judge's marks, the skill hints and the update check move
+  too. So do the lesson-taint index, Cursor's marks and replies, the snapshot
+  and replay indexes and the taste-learn stamps. The first open of the
+  database imports each store's old files once and deletes them. A later open
+  deletes, unread, a file an older release still running writes there. Snapshot
+  blobs, replay runs and logs stay files. `tezgah-doctor --clean` sweeps the
+  stamps and marks there.
+
 ## [1.3.0] - 2026-10-07
 
 ### Added
@@ -833,7 +933,9 @@ adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
   the copy could not find `hooks/` and died on import. It now links into the
   installed tree, and also installs the command as `tezgah`.
 
-[Unreleased]: https://github.com/r1z4x/tezgah/compare/v1.3.0...HEAD
+[Unreleased]: https://github.com/r1z4x/tezgah/compare/v1.5.0...HEAD
+[1.5.0]: https://github.com/r1z4x/tezgah/releases/tag/v1.5.0
+[1.4.0]: https://github.com/r1z4x/tezgah/releases/tag/v1.4.0
 [1.3.0]: https://github.com/r1z4x/tezgah/releases/tag/v1.3.0
 [1.2.0]: https://github.com/r1z4x/tezgah/releases/tag/v1.2.0
 [1.1.0]: https://github.com/r1z4x/tezgah/releases/tag/v1.1.0

package/MANIFEST

@@ -142,6 +142,7 @@ hooks/tezgah_skill_pick.py
 hooks/tezgah_snapshot.py
 hooks/tezgah_stack.py
 hooks/tezgah_stopspec.py
+hooks/tezgah_store.py
 hooks/tezgah_task.py
 hooks/tezgah_taste.py
 hooks/tezgah_taste_ledger.py
@@ -670,6 +671,7 @@ tests/test_ai_research_library.py
 tests/test_apps_registry.py
 tests/test_attest.py
 tests/test_bash_agreement.py
+tests/test_cache_store.py
 tests/test_cheap_source.py
 tests/test_clarity.py
 tests/test_claude_adapters.py
@@ -694,6 +696,7 @@ tests/test_dsh_statusline.py
 tests/test_embed.py
 tests/test_encoding.py
 tests/test_enforcers.py
+tests/test_evidence_store.py
 tests/test_fuzz_shell.py
 tests/test_gate.py
 tests/test_gate_vectors.py

package/agents/tezgah-reviewer.md

@@ -3,7 +3,9 @@ name: tezgah-reviewer
 description: >
   Adversarial, read-only review of a diff, branch or PR: derives the blast
   radius, checks correctness, contract, security, tests and performance, and
-  classifies every finding confirmed/refuted/unverified.
+  classifies every finding confirmed/refuted/unverified. One review per
+  plan, at most two rounds; spawn one per dimension in one message for a
+  large diff.
 model: opus
 effort: high
 readonly: true
@@ -56,15 +58,22 @@ is lowered only by evidence: a check you could not complete never
 downgrades a confirmed finding - keep its severity and mark it
 provisional. Disclose the order you read the files in.
 
-Scope and budget. Every round that confirms a defect is followed by a
-round over that fix's delta; the review ends when a round confirms none.
-A later round reads only the delta since the sha the previous round
-recorded (`tezgah-task review` prints it): re-reading the whole diff
-re-reports what was already triaged. Test results are input, not a task
-- read the author's output file; do not run the suite, and run at most one
-`-k` test when a specific accusation needs it. The same defect family
-confirmed three times is a sign the design needs a pivot, not another
-patch.
+Scope and budget. One review per plan, over its whole diff in the
+verification phase; at most two rounds - only a confirmed critical or
+major finding opens round two, a minor or suggestion one is recorded
+fix-later, and after round two what is left goes to the user or is
+recorded fix-later, never a third round. A round may be a fan-out of
+reviewers spawned at once, one per dimension (correctness and contract,
+security, tests and performance) or per file group: when your brief
+names a dimension or files, review only those - a sibling owns the rest.
+Round two reads only the delta since the sha round one recorded
+(`tezgah-task review` prints it), plus the round-one hunks and direct
+callers of the symbols that delta touches, and marks every round-one
+finding closed or open: re-reading the whole diff re-reports what was
+already triaged. Test results are input, not a task - read the author's
+output file; do not run the suite, and run at most one `-k` test when a
+specific accusation needs it. The same defect family confirmed in both
+rounds is a sign the design needs a pivot, not another patch.
 
 You may use: the codegraph MCP tools, read, grep and glob. Read-only:
 no writes, no edits, no shell.

package/bin/tezgah-docs

@@ -24,7 +24,7 @@ every word is part of the query.
 
 Matching is word-based and case-insensitive against an entry's title, its
 questions, its sources and its path; every word must match, so a second word
-narrows. A query that matches nothing is the last resort of one TypeSafe Choice
+narrows. A query that matches nothing is the last resort of one judge-seam Choice
 question over the pages themselves; it answers `none` for a query no page
 covers. When that question gives no judgement - unavailable (`docs-judge-off`,
 the seam's `judge-off`, no credential) or a call that failed or came back
@@ -106,7 +106,7 @@ RUN = re.compile(r"((?:`[A-Za-z_][A-Za-z0-9_]*(?:\(\))?`\s*[/,]?\s*){1,5})$")
 CITE = re.compile(r"`([\w.-]+(?:/[\w.-]+)+|[\w-]+\.\w+):(\d+)(?:-(\d+))?`"
                   r"|`(:(\d+)(?:-(\d+))?)`")
 # The symbol written after the citation instead of before it, the way a table
-# cell reads: `hooks/tezgah_policy.py:667-928` (`CORE`).
+# cell reads: `hooks/tezgah_policy.py:678-946` (`CORE`).
 AFTER = re.compile(r"\s*\(`([A-Za-z_][A-Za-z0-9_]*)(?:\(\))?`\)")
 # Any `path:line`, backticked or not: what CITE does not match of these (a
 # bare token, a comma list `x.py:12,40`) is counted as unjudged, so a citation
@@ -1036,7 +1036,7 @@ def record(result):
 
 
 def judge_pick(query):
-    """(judged, page): whether a TypeSafe Choice question came back with one of
+    """(judged, page): whether a judge-seam Choice question came back with one of
     its options, and the page it named (None for `none`).
 
     The last resort for a query the index could not place: a word the index does

package/bin/tezgah-doctor

@@ -34,12 +34,14 @@ this tree.
   tezgah-doctor --coverage         the tracked files the graph index does not hold
   tezgah-doctor --coverage --repo ~/Projects/x
   tezgah-doctor --clean            vacuum the opencode database, and delete the
-                                   hook state (evidence ledgers, turns/,
-                                   sessions/, classify.log, context-drops.log,
-                                   debug.log) idle > 30 days
-                                   (--retention-days N); $TEZGAH_SESSION's
-                                   own files are always kept
+                                   hook state (evidence ledgers, turn
+                                   stamps, used-tool marks, classify.log,
+                                   context-drops.log, debug.log) idle > 30
+                                   days (--retention-days N);
+                                   $TEZGAH_SESSION's own state is always kept
   tezgah-doctor --prune-sessions 30   delete sessions idle > 30 days, then vacuum
+  tezgah-doctor --import-evidence  import the legacy JSONL evidence ledgers
+                                   into the evidence database now
 """
 import argparse
 import importlib.machinery
@@ -79,7 +81,11 @@ GRAPH_DB = CODEGRAPH + ".db"
 # The per-session state the hooks write and nothing else removes: the evidence
 # ledgers, the per-turn stamps, the used-tool marks and the two capped logs.
 # Audit L-11 (GAP-10) measured 2,000 of each, 400 days old, surviving a full
-# cycle and `--clean`. Snapshots are not here: tezgah_snapshot caps them itself.
+# cycle and `--clean`. The stamps and the marks are rows of the cache database
+# (`tezgah_store.sweep_cache`); the evidence directory holds the legacy ledger
+# files its import has not taken, and `turns/` and `sessions/` the files a
+# writer from an older release still running beside this one adds there.
+# Snapshots are not here: tezgah_snapshot caps them itself.
 STATE_DIRS = ("evidence", "turns", "sessions")
 STATE_LOGS = ("classify.log", "context-drops.log", "debug.log")
 RETENTION_DAYS = 30
@@ -277,20 +283,36 @@ def prune_sessions(days):
 
 def sweep_state(days, session=None):
     """Delete the hook state not modified for `days`, in the global cache and
-    the temp fallback; (removed, bytes freed).
-
-    By mtime because every writer appends or rewrites in place, so a file's
-    mtime is its session's last activity. The current session's files
-    (`TEZGAH_SESSION`, the id omp and the judgement CLIs export) are kept
-    whatever their age: a session idle past the window and then resumed must
-    not lose its ledger mid-flight."""
+    the temp fallback; (files removed, bytes freed, evidence sessions deleted,
+    evidence rows deleted, turn-stamp and used-mark rows deleted).
+
+    The evidence ledger is swept in its database (`tezgah_store.sweep_evidence`,
+    after importing the legacy JSONL files): every session whose newest row is
+    older than the cutoff loses its rows. The turn stamps and the used-tool
+    marks are swept there too, by the time each was last written
+    (`tezgah_store.sweep_cache`). The other files go by mtime, because
+    every writer appends or rewrites in place, so a file's mtime is its
+    session's last activity. The current session (`TEZGAH_SESSION`, the id omp
+    and the judgement CLIs export) is kept whatever its age: a session idle past
+    the window and then resumed must not lose its ledger mid-flight."""
+    import tezgah_store
     from tezgah_context import slug
     from tezgah_integrity import _slug
     keep = ({_slug(session) + ".jsonl", slug(session) + ".json",
              slug(session) + ".jsonl"} if session else set())
     cutoff = time.time() - days * 86400
-    removed = freed = 0
+    removed = freed = sessions = rows = marks = 0
     for base in dict.fromkeys((CACHE, fallback_cache())):
+        try:
+            swept = tezgah_store.sweep_evidence(base, cutoff,
+                                                _slug(session) if session else None)
+        except tezgah_store.ERRORS:
+            swept = (0, 0)
+        sessions, rows = sessions + swept[0], rows + swept[1]
+        try:
+            marks += tezgah_store.sweep_cache(base, cutoff, slug(session) if session else None)
+        except tezgah_store.ERRORS:
+            pass
         paths = [os.path.join(base, name) for name in STATE_LOGS]
         for sub in STATE_DIRS:
             try:
@@ -310,7 +332,15 @@ def sweep_state(days, session=None):
                 continue
             removed += 1
             freed += st.st_size
-    return removed, freed
+    return removed, freed, sessions, rows, marks
+
+
+def import_evidence():
+    """The bulk import of the legacy JSONL ledgers into the evidence database
+    of the cache and of the temp fallback: {cache: files seen}."""
+    import tezgah_store
+    return {base: tezgah_store.import_evidence(base)
+            for base in dict.fromkeys((CACHE, fallback_cache()))}
 
 
 
@@ -421,7 +451,7 @@ def main():
                     help="a saved `codegraph files -j` dump to read for --coverage")
     ap.add_argument("--clean", action="store_true",
                     help="vacuum the opencode database and delete tezgah hook "
-                         "state (ledgers, turns/, sessions/, classify.log, "
+                         "state (ledgers, turn stamps, used-tool marks, classify.log, "
                          "context-drops.log, debug.log) not "
                          "modified for --retention-days; the current session "
                          "($TEZGAH_SESSION) is kept (opt-in)")
@@ -431,6 +461,10 @@ def main():
                          "%(default)s days)")
     ap.add_argument("--prune-sessions", type=int, metavar="DAYS",
                     help="delete sessions idle longer than DAYS, then vacuum")
+    ap.add_argument("--import-evidence", action="store_true",
+                    help="import the legacy JSONL evidence ledgers into the "
+                         "evidence database now, instead of at the next "
+                         "session start's daily run")
     args = ap.parse_args()
     if args.retention_days < 1:
         ap.error("--retention-days wants a positive day count")
@@ -439,6 +473,19 @@ def main():
         repo = os.path.realpath(os.path.expanduser(args.repo[0]) if args.repo
                                 else os.getcwd())
         return coverage_report(repo, args.dump)
+    if args.import_evidence:
+        import tezgah_store
+        try:
+            seen = import_evidence()
+        except tezgah_store.ERRORS as exc:
+            print("evidence import failed: %s" % exc, file=sys.stderr)
+            return 1
+        if args.json:
+            print(json.dumps({"imported": seen}, indent=2))
+        else:
+            for base, n in seen.items():
+                print("%s: %d legacy ledger file(s) imported" % (base, n))
+        return 0
     if args.stack:
         import tezgah_stack
         repo = os.path.realpath(os.path.expanduser(args.repo[0]) if args.repo
@@ -470,10 +517,13 @@ def main():
                 except sqlite3.Error as exc:
                     cleaned["vacuum"] = "failed: %s" % exc
         if args.clean:
-            removed, freed = sweep_state(args.retention_days,
-                                         os.environ.get("TEZGAH_SESSION"))
+            removed, freed, sessions, rows, marks = sweep_state(
+                args.retention_days, os.environ.get("TEZGAH_SESSION"))
             cleaned["state"] = {"removed": removed, "bytes": freed,
                                 "days": args.retention_days}
+            cleaned["evidence"] = {"sessions": sessions, "rows": rows,
+                                   "days": args.retention_days}
+            cleaned["marks"] = {"rows": marks, "days": args.retention_days}
         rep = collect(pairs)
 
     if args.json:
@@ -499,6 +549,13 @@ def main():
         if st:
             print("  hook state    %d file(s) older than %d days removed, %s"
                   % (st["removed"], st["days"], human(st["bytes"])))
+        ev = cleaned.get("evidence")
+        if ev:
+            print("  evidence      %d session(s) idle > %d days deleted (%d rows)"
+                  % (ev["sessions"], ev["days"], ev["rows"]))
+        mk = cleaned.get("marks")
+        if mk:
+            print("  stamps/marks  %d row(s) idle > %d days deleted" % (mk["rows"], mk["days"]))
     elif rep["opencode_db_bytes"] > 500 * 1024 * 1024:
         print("\nrun with --clean to vacuum; --prune-sessions DAYS to shrink the db")
     return 0

package/bin/tezgah-route

@@ -12,8 +12,11 @@ A brief naming stored data, a migration, credentials, the gate or security goes
 tezgah-frontier by rule; otherwise Jev picks the tier (one Choice, ~650 input
 tokens, ~0.4 s); with no key or `judge-off`, `--phase`
 (explore|code|mechanical|plan|review|research) picks it from the static table, and
-with neither the middle tier is used. The brief is sent to api.typesafe.ai (or the
-OpenRouter fallback) - write it in English, the language the judge reads best.
+with neither the middle tier is used. The brief is sent to the judge seam's
+provider - the session's own CLI (omp, claude, opencode, cursor, codex) or the Jev
+model through TypeSafe, OpenRouter System One, Cloudflare Workers AI or a
+compatible endpoint (TEZGAH_JEV_PROVIDER picks one), else the OpenRouter chat
+fallback - so write it in English, the language the judge reads best.
 
 Every routed call writes one `route` row to the session's ledger (TEZGAH_SESSION):
 tier, agent, what set the tier (`via`: rule|jev|phase|default), the static

package/bin/tezgah-setup

@@ -59,6 +59,7 @@ import tezgah_paths as tp  # noqa: E402
 import tezgah_policy as policy  # noqa: E402
 import tezgah_agents  # noqa: E402
 import tezgah_integrity  # noqa: E402
+import tezgah_store  # noqa: E402
 import tezgah_attest as ta  # noqa: E402
 from tezgah_policy import (  # noqa: E402
     ADHD, CODEGRAPH_RULE, CODEGRAPH_STATIC, CONSULT, EXEC, NO_CONSULT,
@@ -883,6 +884,22 @@ def save_features(ids):
     return True
 
 
+def set_jev(value):
+    """Persist `value` as config.json["jev"] - this key's only writer - and print
+    the carriers it resolves to now. An unchanged value writes nothing, like
+    `save_features`; a config that does not parse is refused, not replaced."""
+    import tezgah_judge
+    cfg = read_config(tp.CONFIG, {})
+    if cfg is None:
+        sys.stderr.write("tezgah-setup: %s is not valid JSON; jev not set\n" % tp.CONFIG)
+        return 1
+    if cfg.get("jev") != value:
+        cfg["jev"] = value
+        write_json(tp.CONFIG, cfg)
+    print("%s %s  %s" % (OK, tp.CONFIG, tezgah_judge.jev_summary()))
+    return 0
+
+
 def app_servers(selected=None):
     """The app-analysis MCP servers to wire this run (shared spec)."""
     return tezgah_apps.servers(selection() if selected is None else selected)
@@ -4161,17 +4178,6 @@ def _denied(host, stdout):
     return (got.get("hookSpecificOutput") or {}).get("permissionDecision") == "deny"
 
 
-def _live_cleanup(slug):
-    """Remove every cache file the synthetic session left (named by its slug)."""
-    for dirpath, _dirs, files in os.walk(tp.cache_dir()):
-        for name in files:
-            if slug in name:
-                try:
-                    os.remove(os.path.join(dirpath, name))
-                except OSError:
-                    pass
-
-
 def live_probe(host, deny=False):
     """(state, line) for one host: ok, miss or unverified. `deny` sends the
     PreToolUse probe: the host's deny envelope and a deny row must both appear;
@@ -4211,15 +4217,14 @@ def live_probe(host, deny=False):
                                   capture_output=True, text=True, env=env,
                                   cwd=root, timeout=LIVE_TIMEOUT)
         ms = int((time.monotonic() - started) * 1000)
-        try:
-            with open(ledger, encoding="utf-8") as fh:
-                kinds = [json.loads(ln).get("kind") for ln in fh if ln.strip()]
-        except (OSError, ValueError):
-            kinds = []
+        kinds = [row.get("kind") for row in tezgah_integrity.events_path(ledger)]
     except (OSError, subprocess.SubprocessError) as exc:
         return "miss", "live %s: the hook did not run (%s)" % (who, exc)
     finally:
-        _live_cleanup(slug)
+        try:
+            tezgah_store.forget_session(ledger)
+        except tezgah_store.ERRORS:
+            pass
         shutil.rmtree(root, ignore_errors=True)
     err = (proc.stderr or "").strip().splitlines()
     tail = " (exit %d%s)" % (proc.returncode, ": " + err[-1][:160] if err else "")
@@ -4778,7 +4783,7 @@ def host_checks_omp():
          tp.have_typesafe_key()),
         # One row per channel: the two disagree in both directions - a key file
         # alone serves the seam, an omp login-store record alone serves omp.
-        ("judgement seam key resolvable (env or key file)",
+        ("judgement seam key resolvable (session CLI, Jev carrier or OpenRouter)",
          tp.have_judge_key()),
         ("status line answers", omp_hook_answers()), ("tier models in task.agentModelOverrides and modelRoles.plan/slow", tier_models_current()),
     ]
@@ -5690,6 +5695,11 @@ def main(argv):
                     help="with --install: the language replies are asked for and "
                          "the Stop rule checks (config.json `reply_lang`; `tr` "
                          "when unset, `any` checks none)")
+    ap.add_argument("--jev", choices=tp.JEV_CHOICES,
+                    help="the carrier the judgement seam reaches the Jev model "
+                         "through (config.json `jev`; `auto` tries every one "
+                         "that resolves); TEZGAH_JEV_PROVIDER and JEV_PROVIDER "
+                         "still win over it")
     # `--version` is argparse's print-and-exit action, so `--upgrade --version V`
     # would print the current version, exit 0 and move nothing at all - the
     # "reads as done, did nothing" shape this repo hunts. Refused by name here,
@@ -5742,6 +5752,8 @@ def main(argv):
     if args.features:
         features_report()
         return 0
+    if args.jev:
+        return set_jev(args.jev)
     if args.refresh:
         paths = refresh_contract()
         for path in paths:

package/bin/tezgah-status

@@ -298,7 +298,9 @@ def main():
         last = tj.last_use()
         if "--json" in flags:
             print(json.dumps(last, sort_keys=True))
-        elif not last:
+            return 0
+        print(tj.jev_summary())
+        if not last:
             print("judge: no judgement recorded")
         elif last.get("provider"):
             print("judge: %s answered with %s/%s (fallback=%s)%s" % (