@r1z4x/tezgah 1.4.0 → 1.5.1

package/CHANGELOG.md

@@ -6,6 +6,123 @@ adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ## [Unreleased]
 
+## [1.5.1] - 2026-10-09
+
+### Fixed
+
+- **Subagents load in the first session.** Claude and Codex missed every role a
+  session wrote into a new repo until the next session. Codex also ignored
+  `.codex/agents` in an untrusted repo. The install now writes the roles to
+  `~/.claude/agents` and `$CODEX_HOME/agents`, as it already did for omp. A
+  session removes the copies an older release left in a repo's `.claude/agents`
+  and `.codex/agents`. Cursor read 9 of 10 role descriptions as ">" and got a
+  model id it does not list. It now gets its own `.cursor/agents` with a
+  one-line description and `model: inherit`.
+
+- **`tezgah-route` names the specialists.** It could only name a tier. The
+  five specialists ran 0 times in 121 delegations. The generic task agent ran 86.
+  The judge now also answers whether a brief is a specialist's whole
+  job, and the route names that agent. `--phase review` and `--phase research`
+  name the reviewer and the researcher. A security brief that only reviews
+  goes to `tezgah-security`. The routing and fan-out text say to spawn the
+  named agent, never the generic one.
+- **Jev through OpenRouter asks a model OpenRouter has.** The carrier sent
+  `typesafe/jev-latest`, which OpenRouter answers with 400 (no such model). It
+  now sends the bare `jev-latest`, which OpenRouter maps itself.
+- **The Workers AI model id picks Clef.** `tezgah-setup` refused
+  `--jev @cf/cloudflare/clef-flash`, and the same value in
+  `TEZGAH_JEV_PROVIDER` chose no carrier. Both now name `clef-flash`, and
+  `@cf/cloudflare/clef` names `clef`.
+- **opencode's judge run denies tools at agent level too.** A user's
+  `agent.build.permission` beat the inline top-level deny.
+- **The refusal note needs an asked session CLI.** The seam writes it only
+  when it asked the session CLI first. A caller that named the CLI gets it.
+- **A chosen carrier reads only its own key.** A `clef` choice no longer opens
+  omp's login database on every status-line redraw.
+- **`pnpm test` names a command.** An Acceptance item counted as a command
+  only when a word after the first was an option or a path. So `pnpm test`,
+  `cargo test` and `make test` read as missing, and the phase move refused the
+  plan. A known runner followed by `test` or `run` now counts. Prose such as
+  `run the tests` still does not.
+- **Each carrier's price is now pinned by a test.** The one cost check priced a
+  TypeSafe answer at Jev's rate. It passed with or without per-carrier pricing.
+  New tests in `tezgah-taste`, `tezgah-triage` and the replay labeller price Clef
+  and Clef-flash answers at their own rates. A run with an unpriced answer prints
+  `cost unknown`.
+- **Claude lists the reviewer once.** The plugin shipped its own
+  `agents/tezgah-reviewer.md`, which Claude shows as `tezgah:tezgah-reviewer`
+  beside the generated reviewer and never shadows. The plugin now ships no
+  agent, and its release-time renderer is gone with it.
+
+## [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
@@ -864,7 +981,9 @@ 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/v1.4.0...HEAD
+[Unreleased]: https://github.com/r1z4x/tezgah/compare/v1.5.1...HEAD
+[1.5.1]: https://github.com/r1z4x/tezgah/releases/tag/v1.5.1
+[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

package/MANIFEST

@@ -19,7 +19,6 @@ README.md
 README.tr.md
 RELEASING.md
 SECURITY.md
-agents/tezgah-reviewer.md
 assets/logo/README.md
 assets/logo/tezgah-logo-16.png
 assets/logo/tezgah-logo-256.png

package/README.md

@@ -109,10 +109,12 @@ it set out to arm is not armed.
   `~/.codex/AGENTS.md`, `~/.omp/agent/RULES.md`). Claude Code's hooks ship in a
   plugin registered through `claude plugin`; with no `claude` CLI, Claude stays
   unarmed and the report says so.
-- **Agent files in your repositories.** Inside the configured roots, a session
-  writes subagent files to `.claude/agents/`, `.codex/agents/` and
-  `.opencode/agents/`, and lists those directories in the clone's own
-  `.git/info/exclude` (`TEZGAH_NO_EXCLUDE=1` stops that).
+- **Agent files.** The install writes tezgah's subagent roles to
+  `~/.claude/agents/`, `$CODEX_HOME/agents/` (default `~/.codex/agents/`) and
+  `~/.omp/agent/agents/`. Inside the configured roots, a session writes them to
+  the repository's `.cursor/agents/` and `.opencode/agents/`, and lists those
+  directories in the clone's own `.git/info/exclude` (`TEZGAH_NO_EXCLUDE=1`
+  stops that).
 - **A daily update check.** The status line asks for a newer release at most
   once a day; `~/.config/tezgah/update-check-off` or `TEZGAH_UPDATE_CHECK=0`
   turns it off.

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:680-948` (`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-route

@@ -1,5 +1,5 @@
 #!/usr/bin/env python3
-"""tezgah-route - which tier worker a delegated task goes to, and the model table.
+"""tezgah-route - which worker a delegated task goes to, and the model table.
 
   tezgah-route "<brief>" [--phase P] [--json]   the agent to spawn, and why
   tezgah-route --check                          the table's age and refresh flags
@@ -9,11 +9,17 @@
   tezgah-route --report                         each route's first check, by tier
 
 A brief naming stored data, a migration, credentials, the gate or security goes to
-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.
+tezgah-frontier by rule - or to tezgah-security when it only reviews; otherwise
+Jev answers two Choices in one call (~700 input tokens, ~0.4 s): the tier, and
+whether the brief is the whole job of a specialist (tezgah-tester, -docs,
+-security, -ui, -researcher, -reviewer), which is then the agent named. With no
+key or `judge-off`, `--phase review|research` names the reviewer or the
+researcher, another `--phase` (explore|code|mechanical|plan) picks the tier from
+the static table, and 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

@@ -13,7 +13,6 @@
   tezgah-setup --adopt                 move the predecessor wiring aside
   tezgah-setup --sync                  dev: copy this checkout over the Claude plugin copy
   tezgah-setup --write-manifest        release: regenerate the tracked MANIFEST from git
-  tezgah-setup --write-plugin-agents   release: re-render the plugin's agents/ from the role source
   tezgah-setup --status [PATH]         print the armed/used checklist
   tezgah-setup --agents [PATH]         regenerate this repo's subagent set
   tezgah-setup --deps [--dry-run]      install missing optional tools only
@@ -884,6 +883,22 @@ 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)
@@ -1147,9 +1162,33 @@ def install_claude():
     # (`TEZGAH_CORE_IN_FILE`, hooks/projects-auto-init.py), so a session pays for
     # the contract once.
     print("%s %s (always-on contract)" % (OK, write_rules("claude")))
+    write_user_agents("claude")
     register_claude_plugin()
 
 
+def write_user_agents(host):
+    """The host's tezgah roles in its user agent dir, where it loads them from
+    the first session (a repo dir written at session start is missed there),
+    then tezgah's MARKER files no longer wanted - `agents-off`, a lost
+    capability, a retired role - are removed. The user's own files stay."""
+    agents_dir = tezgah_agents.user_agent_dir(host)
+    wanted = tezgah_agents.user_agents(host)
+    for name, text in sorted(wanted.items()):
+        path = os.path.join(agents_dir, name)
+        if write_if_changed(path, text):
+            print("%s %s" % (OK, path))
+    for path in tezgah_agents.remove_stale(agents_dir, wanted):
+        print("%s removed %s" % (OK, path))
+
+
+def user_agent_files(host):
+    """{filename: text} of tezgah's MARKER files in the host's user agent dir."""
+    d = tezgah_agents.user_agent_dir(host)
+    return {n: t for n in (os.listdir(d) if os.path.isdir(d) else [])
+            if n.startswith("tezgah-")
+            and tezgah_agents.MARKER in (t := tezgah_agents._read(os.path.join(d, n)) or "")}
+
+
 # The marketplace a manifest rendered by this installer carries, so Claude
 # installs the plugin as `tezgah@tezgah-local`. A manifest the tree already has
 # under another name (the maintainer's) is kept, name included.
@@ -1342,6 +1381,7 @@ def install_codex():
     # instructions Codex keeps for the whole session. Global, like omp's
     # RULES.md, because a per-project file would be a tracked project file.
     print("%s %s (always-on contract)" % (OK, write_rules("codex")))
+    write_user_agents("codex")
 
 
 def _toml_str(value):
@@ -3277,6 +3317,9 @@ def verify_uninstall(hosts, full):
         # the user's own CLAUDE.md may name tezgah; only the block is ours
         clean("claude global CLAUDE.md block gone", [CLAUDE_RULES],
               needle="<!-- tezgah:start")
+        claim("claude subagents gone",
+              [os.path.join(tezgah_agents.user_agent_dir("claude"), n)
+               for n in user_agent_files("claude")])
     if "codex" in hosts:
         clean("codex hooks.json unwired", [os.path.join(CODEX, "hooks.json")])
         claim("codex skills unlinked",
@@ -3288,6 +3331,9 @@ def verify_uninstall(hosts, full):
         clean("codex global AGENTS.md block gone",
               [os.path.join(CODEX, n) for n in CODEX_RULES],
               needle="<!-- tezgah:start")
+        claim("codex subagents gone",
+              [os.path.join(tezgah_agents.user_agent_dir("codex"), n)
+               for n in user_agent_files("codex")])
     if "opencode" in hosts:
         claim("opencode plugin unlinked",
               [os.path.join(OPENCODE, d, "tezgah.js") for d in ("plugins", "plugin")]
@@ -3438,6 +3484,13 @@ def uninstall_claude():
     unwire_claude_settings()
     drop_claude_mcp_row()
     strip_managed_md(CLAUDE_RULES)
+    remove_user_agents("claude")
+
+
+def remove_user_agents(host):
+    """Every tezgah MARKER file in the host's user agent dir; the user's stay."""
+    for path in tezgah_agents.remove_stale(tezgah_agents.user_agent_dir(host), ()):
+        print("%s removed %s" % (OK, path))
 
 
 def drop_claude_mcp_row():
@@ -3472,6 +3525,7 @@ def uninstall_codex():
     # both names: the override may have appeared, or gone, since the install
     for name in CODEX_RULES:
         strip_managed_md(os.path.join(CODEX, name))
+    remove_user_agents("codex")
 
 
 def uninstall_opencode():
@@ -4413,6 +4467,8 @@ def host_checks_claude():
          bool(copies) and all(plugin_copy_current(t) for t in copies)),
         ("CLAUDE.md carries the contract (matches a fresh render)",
          rules_current("claude")),
+        ("subagents in %s (match a fresh render)" % tezgah_agents.user_agent_dir("claude"),
+         user_agent_files("claude") == tezgah_agents.user_agents("claude")),
     ] + claude_mcp_rows(copies)
 
 
@@ -4601,6 +4657,8 @@ def host_checks_codex():
         ("MCP in config.toml", "mcp_servers.codegraph" in toml),
         ("%s carries the contract (matches a fresh render)"
          % os.path.basename(codex_rules()), rules_current("codex")),
+        ("subagents in %s (match a fresh render)" % tezgah_agents.user_agent_dir("codex"),
+         user_agent_files("codex") == tezgah_agents.user_agents("codex")),
     ] + mcp_rows(toml_servers(toml_path), "config.toml") + [
         mcp_row(TEZGAH_MCP_NAME, toml_mcp_entry(toml_path, TEZGAH_MCP_NAME),
                 "config.toml", tezgah_mcp_command()),
@@ -4767,7 +4825,7 @@ 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()),
     ]
@@ -4913,24 +4971,6 @@ def write_manifest():
     return 0
 
 
-def write_plugin_agents():
-    """Re-render the plugin's tracked `agents/*.md` from tezgah_agents' role
-    bodies, the source the per-repo agents are generated from. A plain write,
-    not write_if_changed: a tracked file needs no `.tezgah-bak` beside it."""
-    for name, text in tezgah_agents.plugin_agents().items():
-        path = os.path.join(HERE, "agents", name)
-        try:
-            with open(path, encoding="utf-8") as fh:
-                if fh.read() == text:
-                    continue
-        except OSError:
-            pass
-        with open(path, "w", encoding="utf-8") as fh:
-            fh.write(text)
-        print("%s %s" % (OK, path))
-    return 0
-
-
 def tree_sha(root, files):
     """sha256 over each relpath and its bytes, or None when one is unreadable."""
     h = hashlib.sha256()
@@ -5631,8 +5671,6 @@ def main(argv):
                     help="regenerate this repo's subagent set (default: cwd)")
     ap.add_argument("--write-manifest", action="store_true",
                     help="release: regenerate the tracked MANIFEST from git")
-    ap.add_argument("--write-plugin-agents", action="store_true",
-                    help="release: re-render the plugin's agents/ from the role source")
     ap.add_argument("--deps", action="store_true")
     ap.add_argument("--mcp-schemas", action="store_true",
                     help="measure the MCP tool-schema band by asking each server")
@@ -5679,6 +5717,11 @@ 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", type=tp.jev_name, 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,
@@ -5731,6 +5774,8 @@ 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:
@@ -5740,8 +5785,6 @@ def main(argv):
         return 0
     if args.write_manifest:
         return write_manifest()
-    if args.write_plugin_agents:
-        return write_plugin_agents()
     if args.status:
         print(tezgah_context.health_lines(os.path.realpath(expand(args.status))))
         # The install-health half of this report, from the same function the

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" % (