@dzhechkov/harness-cli 0.4.1 → 0.4.3

package/.dz-manifest.json

@@ -9,7 +9,7 @@
       },
       {
         "path": "README.md",
-        "sha256": "357bb891e287a2af120a66914085b3b7dec24092aa99f46a8767c096741d713f"
+        "sha256": "532051106b153c0d9bb7ad2abbeb71d418f88f1bc88e60a35e29cf42162af243"
       },
       {
         "path": "coverage/coverage-final.json",
@@ -37,15 +37,15 @@
       },
       {
         "path": "dist/cli.d.ts.map",
-        "sha256": "a8cff82f9f31b06b6a5f1934747631fcc551945ee3b7ac2c620b7c25076303cc"
+        "sha256": "5bdc153e789ac180543e9b4955ea36666dc27b7238637b05875e048fc49ae30f"
       },
       {
         "path": "dist/cli.js",
-        "sha256": "7afd2be99a3b9581b6fd8bfae58df6c608db346f741263c219d2140a5e8300ce"
+        "sha256": "0183c01a98e232554a7460be48ee5fdf59dd3a7d09c719a31b08acc9c6b3528c"
       },
       {
         "path": "dist/cli.js.map",
-        "sha256": "481e6ddd256869585b3450ebe901057398a30bb8292e6f93f80a18d5c89ad0c0"
+        "sha256": "c5bd1b0b852dd21de328494d7a99db336afbde4397b7c687d416f1a6f3c297b2"
       },
       {
         "path": "dist/index.d.ts",
@@ -69,7 +69,7 @@
       },
       {
         "path": "package.json",
-        "sha256": "5a2d203ef20d2b6179595627b4fa1856fd7e17a31b19690324f589487874f9e0"
+        "sha256": "4a164e5ac53ceec178463c1bdab9f80c1755311d5e4202c1f6234ca22c473b37"
       },
       {
         "path": "src/bin.ts",
@@ -77,7 +77,7 @@
       },
       {
         "path": "src/cli.ts",
-        "sha256": "1aba996c4e5dd717a496074baa83778e8f2e04927689fc2cfc315033b8982b38"
+        "sha256": "0a1097fc3a2294f934fc50f5871b287cfbc5ec85becdab914b7caeca42cb7d1a"
       },
       {
         "path": "src/index.ts",
@@ -85,7 +85,7 @@
       },
       {
         "path": "test/cli.test.ts",
-        "sha256": "e63d24934506b3d6c6ec7f67072273ed9c9dabb86e8a24a7838819217ca6035d"
+        "sha256": "c66bdd2fbbe2e0b622e3bc4e0e4521e865cc8088e581bd4b4862c1c550307900"
       },
       {
         "path": "test/epoch-replay-cli.test.ts",
@@ -189,7 +189,15 @@
       },
       {
         "path": "test/mutation-registry.json",
-        "sha256": "36276fa51f9e5301ddb317687d0dbc839641b86f47f203e5d392bab31f4f178e"
+        "sha256": "fe8e5792626cf58678c1e2bcb809f63413655c2e171439157b74deb8b4fd80fb"
+      },
+      {
+        "path": "test/statusline-panel.test.ts",
+        "sha256": "b09b42c4ece1a83e5d3f3765629b0756a8f30a7122f3fc30363f93f092a44c0d"
+      },
+      {
+        "path": "test/trace-bundle-cli.test.ts",
+        "sha256": "fdaad63359ca7ce351e505e613b547bbc653e0847403edfae88d22a8a524b7a0"
       },
       {
         "path": "test/workflow-init-lint-clean.test.ts",
@@ -213,5 +221,5 @@
       }
     ]
   },
-  "signature": "YV90oAIqb+16MFtctEUA4hH5x/O8BbGdAGh+eQ4HWJSxDwiRG3NxBDbDXFlUBvWKxwT/ttmcgKgaEE1ZHeIyDg=="
+  "signature": "wIo2tChkUqPT1qEHFeLlgbuzL29rRhhpVer4yaDmeuq7CsvdMeYN2QcttC496DwlepKeZ1EE3h2sSoY5w16qAA=="
 }

package/README.md

@@ -1141,6 +1141,57 @@ host run there is nothing to read, and it says so rather than inventing a timeli
 follows from the same boundary: on a non-Claude-Code target the authoring and lint verbs work
 unchanged, and only execution is absent.
 
+### Move a run's telemetry to another machine (`workflow-trace export` / `import`)
+
+A run leaves traces on the machine that produced it. `export` puts one run's telemetry into a single
+movable file; `import` reconstructs that run under a root you name.
+
+```bash
+dz workflow-trace export --slug my-feature --o my-feature.bundle.json
+dz workflow-trace import my-feature.bundle.json --into /other/project
+```
+
+**What a bundle carries** — the raw event lines (`trace.jsonl`, `.fa-state/checkpoints.jsonl`), the
+ledger rows selected for that run, optionally the training pairs, and `runMeta`: WHO ran each stage,
+read from the harness's own workflow records. **Events, never aggregates.** The one derived value —
+`attribution`, "which model ran which stage" — travels ALONGSIDE the records it was folded from,
+marked `derived`, naming its rule and the record ids, so a consumer that disagrees can recompute it.
+The rule is stated rather than implied: last-writer-wins by timestamp is a CHOICE — a run whose
+phases used different models has no single honest answer, and the map reports who ran it *last*.
+
+**The ledger selector** matches a row by `runId`, or by `slug` when the row has no `runId` — because
+only `loop-run` rows carry a `runId`, so a `runId`-only filter would select nothing for a feature-adr
+run. The bundle reports rows scanned vs matched, so an empty slice is visibly empty rather than
+indistinguishable from an absent ledger.
+
+**Consent does not travel inside the bundle.** Training pairs may contain target-repo code, so
+including them needs `--include-pairs --yes` at export AND `--with-pairs` at import; a pairs-bearing
+bundle imported without the flag writes no pair content.
+
+**Import is fail-closed.** It reconstructs the run's native layout under `--into`, and REFUSES to
+write into a run directory that already has content unless it is the bundle's own run and `--force`
+is given; an identity mismatch refuses even under `--force`. A refused import writes nothing — not
+one file.
+
+**Degradation is loud and typed**, and exactly one reason asks for action:
+
+| reason | meaning | action |
+|---|---|---|
+| `records-absent` / `no-match` | no harness records, or none for this run | none |
+| `predates-model-routing` | a genuine older run, from before per-stage model routing | none — this is history |
+| `unreadable` | a record could not be parsed | look at that record |
+| `layout-unrecognised` | records exist and parse, but the fields we read are gone | **the harness record layout CHANGED — update the reader** |
+
+By default a degraded export still succeeds and prints one named line per degraded member; `--strict`
+makes it exit non-zero so automation fails closed. The split exists because the actionable reason
+used to fire on normal data — three of thirty-two runs in a real store were simply older than
+per-stage routing — and an alarm that sounds on normal operation stops being an alarm.
+
+**Honest scope:** `runMeta` is read from a store this project does not own, so its shape can change
+without notice. That is precisely what `layout-unrecognised` exists to announce, and why the reader
+refuses rather than half-parsing: a partially-read record would report a model-blind run as
+model-known.
+
 **The v1 plan surface is deliberately NARROW and fully enacted** — `dz workflow validate` REJECTS
 (named diagnostics, never a silent no-op) anything the generated loop would not perform: retry
 timing (`initialDelayMs`/`backoffMultiplier`/`maxDelayMs`/`jitter` — v1 retries are immediate;
@@ -1245,6 +1296,8 @@ dz scout             [--topics <list>] [--since <date>] [--deep] [--output <file
 dz workflow          init --name <n> [--pattern pipeline|barrier|fanout|gate] [--o <plan.json>] | validate <plan.json> [--json] | render <plan.json> --o <script.js> [--check] [--force] | blobs [--check]   # loop-plan/1 authoring (the ADR-005 templates are retired)
 dz workflow-lint     <script.js> [--plan <plan.json>] [--require-plan|--legacy] [--json]   # 17-rule deterministic gate; exit 0/1/3 — inconclusive is never a pass
 dz workflow-trace    <runDir|--slug <s>|--run <id>> [--invariants <plan.json>] [--html <out.html>] [--json]   # timeline + SEQ invariant runner over the loop's own trace.jsonl
+dz workflow-trace export <run> --o <file> [--include-pairs --yes] [--strict]   # one run's telemetry as ONE movable file
+dz workflow-trace import <bundle> --into <root> [--force] [--with-pairs]       # reconstruct that run; fail-closed against clobbering
 dz plugin            [--version <ver>]
 dz downloads
 dz migrate           [--project <dir>]
@@ -1303,7 +1356,46 @@ distills new learnings. It's **opt-in** and **non-clobbering**: `--install` merg
 existing `statusLine` config rather than overwriting it, and nothing changes until you run it.
 Modeled on the Agentic QE statusline pattern (e.g. AQE's `🎓 12 patterns`).
 
+#### Live run segment — `dz statusline --fa-record`
+
+A long-running pipeline can PREPEND its own segment to that line, so the bar shows what is in
+flight rather than only the standing pattern count. The `/feature-adr` pipeline records its
+Pattern-memory state at each step; a generated `loop-designer` loop records its current step on
+every trace flush:
+
+```bash
+# the /feature-adr pipeline — reports the learning loop it is actually running
+dz statusline --fa-record --slug add-user-auth --step "Step 8 QE" --recalled 7 --stored 2 --mode full
+
+# a generated loop — same file, different producer, so it says so
+dz statusline --fa-record --slug my-loop --step build --kind loop
+```
+
+```
+📐 feature-adr Step 8 QE · 🎓 224 pool · ↑7 used · +2 new · ↻0 reinforced · 🎓 dz: 224 patterns · 🧠 3 sources
+🔁 loop build · 🎓 dz: 224 patterns · 🧠 3 sources
+```
+
+`--kind <feature-adr|loop>` names the PRODUCER and defaults to `feature-adr`; an unrecognised value
+exits 1 rather than silently weakening panel arbitration. It is load-bearing in two places:
+
+- **the label**: a loop renders `🔁 loop <step>` and shows **no** learning counters, because a loop
+  never populates them — a `↑0 used` there would be a false statement about the Pattern-memory loop
+  rather than a measurement of it;
+- **the arbitration**: each producer writes its own per-slug slot under
+  `.dz/feature-adr/learning-state/`, and a fresh `feature-adr` state outranks any `loop` state, so a
+  loop running alongside a live pipeline can no longer blank that pipeline's counters. Slots are
+  discovered newest-first, nothing older than 30 minutes is surfaced, and the write path prunes at
+  24 h. Use `--project <dir>` to pin the panel to a specific project root.
+
 ### Usage estimate (`dz usage`)
+> **Pin the weekly reset to an ABSOLUTE instant.** `weeklyResetAnchor: "Wed 08:59"` is
+> server-timezone-relative — measured: the same moment lands a week apart under UTC vs `+03:00`, so
+> after a real account reset the counter can keep showing the OLD week for hours while printing the
+> "correct" clock time. Add your offset: `"Wed 08:59 +03:00"` in `.dz/config.json` — the boundary
+> then never moves with the machine's timezone, and `dz usage` prints the full anchor
+> (`resets Wed 08:59 +03:00`). Without an offset it warns on every run.
+
 
 `dz usage` prints a READONLY, never-throw ESTIMATE of Claude SESSION and WEEKLY token usage,
 aggregated from your local `~/.claude/projects/**/*.jsonl` transcripts. Weekly counts start at the
@@ -1681,6 +1773,27 @@ dz backlog ship 268d3cb1 --reason "shipped in harness-core 0.3.151"
 #   → dz backlog ship: 268d3cb1… new → shipped  add a compounding metric …
 #   (shipping an already-shipped idea is a SAID no-op, exit 0 — safe in cleanup batches)
 dz backlog drop 9286f5eb --reason "superseded by 268d3cb1"    # retire without shipping (→ dropped)
+dz backlog edit 9286f5eb --text "corrected wording"           # rewrite ONE idea's text; every other field
+
+# feed the learned auto-cost routing from REAL run telemetry, then read its advice
+dz routing recommend                # per-stage args.models suggestion, printed WITH its basis:
+                                    # n runs, the time window, the grade-floor rule (success ⇔ QE grade ≥ B,
+                                    # attributed run-level — an inference, and it says so), and every skipped
+                                    # record with WHY. qe is FORCED to the cross-family of the code pick —
+                                    # a same-family qe recommendation is unrepresentable, not filtered.
+dz routing recommend --apply        # feed the samples into .dz/routing-outcomes.json (the store the
+                                    # `auto-cost` plan spec reads) — idempotent by runId: a second apply
+                                    # feeds 0 and names the skipped runs. Insufficient data is SAID
+                                    # (cold-start pick + escalation chain), never dressed as a bar-met pick.
+                                                              # (status, effort, goal, uses…) is preserved byte-for-byte,
+                                                              # the previous text lands in .dz/backlog/edits.jsonl, and the
+                                                              # dedup vector is re-embedded in the same bounded form.
+                                                              # If the re-embed FAILS the edit still lands, exits 1, and the
+                                                              # record is MARKED embedStale — dedup then refuses to trust its
+                                                              # similarity (exact-text identity still applies) until
+                                                              # `dz vector reindex` repairs it. The guard sits where the harm
+                                                              # would be (the future duplicate verdict), not in a warning
+                                                              # nobody re-reads. --append adds instead of replacing; --dry-run previews.
 dz backlog reopen 268d3cb1                                    # changed your mind → back to the pool (new)
 #   reopen on an already-new idea is REFUSED (exit 1) — that is almost always the wrong id.
 #   Terminal→terminal never happens silently: ship on a dropped idea (or drop on shipped) is refused;
@@ -3190,7 +3303,7 @@ npx @dzhechkov/p-replicator init
 
 ## Status
 
-`v0.4.1` — published on npm. Also available as [Claude Plugin](#claude-plugin). Part of [DZ Harness Hub](https://github.com/djd1m/dz-harness-hub).
+`v0.4.3` — published on npm. Also available as [Claude Plugin](#claude-plugin). Part of [DZ Harness Hub](https://github.com/djd1m/dz-harness-hub).
 
 ## Claude Plugin
 

package/dist/cli.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"cli.d.ts","sourceRoot":"","sources":["../src/cli.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAkbH,2EAA2E;AAC3E,MAAM,WAAW,KAAK;IACpB,QAAQ,CAAC,GAAG,CAAC,EAAE,MAAM,CAAC;IACtB,QAAQ,CAAC,KAAK,CAAC,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,IAAI,CAAC;IACxC;;;;OAIG;IACH,QAAQ,CAAC,KAAK,CAAC,EAAE,MAAM,CAAC;IACxB;;;;;OAKG;IACH,QAAQ,CAAC,aAAa,CAAC,EAAE,iBAAiB,CAAC;IAC3C;;;;;OAKG;IACH,QAAQ,CAAC,aAAa,CAAC,EAAE,CAAC,OAAO,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,KAAK,IAAI,CAAC;CACjE;AAED,yFAAyF;AACzF,MAAM,MAAM,iBAAiB,GAAG,CAC9B,GAAG,EAAE,MAAM,EACX,IAAI,EAAE;IAAE,QAAQ,CAAC,GAAG,EAAE,MAAM,CAAC;IAAC,QAAQ,CAAC,SAAS,EAAE,MAAM,CAAA;CAAE,KACvD;IAAE,QAAQ,EAAE,MAAM,CAAC;IAAC,MAAM,EAAE,MAAM,CAAC;IAAC,MAAM,EAAE,MAAM,CAAC;IAAC,QAAQ,CAAC,EAAE,OAAO,CAAA;CAAE,CAAC;AA2xQ9E,wBAAsB,MAAM,CAAC,IAAI,EAAE,MAAM,EAAE,EAAE,EAAE,GAAE,KAAU,GAAG,OAAO,CAAC,MAAM,CAAC,CAoK5E"}
+{"version":3,"file":"cli.d.ts","sourceRoot":"","sources":["../src/cli.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAocH,2EAA2E;AAC3E,MAAM,WAAW,KAAK;IACpB,QAAQ,CAAC,GAAG,CAAC,EAAE,MAAM,CAAC;IACtB,QAAQ,CAAC,KAAK,CAAC,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,IAAI,CAAC;IACxC;;;;OAIG;IACH,QAAQ,CAAC,KAAK,CAAC,EAAE,MAAM,CAAC;IACxB;;;;;OAKG;IACH,QAAQ,CAAC,aAAa,CAAC,EAAE,iBAAiB,CAAC;IAC3C;;;;;OAKG;IACH,QAAQ,CAAC,aAAa,CAAC,EAAE,CAAC,OAAO,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,KAAK,IAAI,CAAC;CACjE;AAED,yFAAyF;AACzF,MAAM,MAAM,iBAAiB,GAAG,CAC9B,GAAG,EAAE,MAAM,EACX,IAAI,EAAE;IAAE,QAAQ,CAAC,GAAG,EAAE,MAAM,CAAC;IAAC,QAAQ,CAAC,SAAS,EAAE,MAAM,CAAA;CAAE,KACvD;IAAE,QAAQ,EAAE,MAAM,CAAC;IAAC,MAAM,EAAE,MAAM,CAAC;IAAC,MAAM,EAAE,MAAM,CAAC;IAAC,QAAQ,CAAC,EAAE,OAAO,CAAA;CAAE,CAAC;AA0lR9E,wBAAsB,MAAM,CAAC,IAAI,EAAE,MAAM,EAAE,EAAE,EAAE,GAAE,KAAU,GAAG,OAAO,CAAC,MAAM,CAAC,CAoK5E"}