@hunter-harness/workflow-harness 0.2.79 → 0.2.81
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/harness/bundles/general/claude-code/.harness-build.json +1 -1
- package/harness/bundles/general/claude-code/harness-archive/SKILL.md +24 -3
- package/harness/bundles/general/claude-code/harness-codebase-map/SKILL.md +13 -3
- package/harness/bundles/general/claude-code/harness-knowledge-ingest/SKILL.md +13 -3
- package/harness/bundles/general/claude-code/harness-knowledge-query/SKILL.md +13 -3
- package/harness/bundles/general/claude-code/harness-plan/SKILL.md +13 -3
- package/harness/bundles/general/claude-code/harness-pull/SKILL.md +1 -1
- package/harness/bundles/general/claude-code/harness-push/SKILL.md +15 -1
- package/harness/bundles/general/claude-code/harness-review/SKILL.md +13 -3
- package/harness/bundles/general/claude-code/harness-run/SKILL.md +13 -3
- package/harness/bundles/general/claude-code/harness-submit/SKILL.md +13 -3
- package/harness/bundles/general/claude-code/harness-sync/SKILL.md +1 -1
- package/harness/bundles/general/claude-code/harness-test/SKILL.md +13 -3
- package/harness/bundles/general/claude-code/scripts/harness_archive.py +158 -44
- package/harness/bundles/general/claude-code/scripts/harness_context.py +77 -3
- package/harness/bundles/general/claude-code/scripts/harness_events.py +35 -2
- package/harness/bundles/general/claude-code/scripts/harness_plan_finalize.py +6 -1
- package/harness/bundles/general/codebuddy/.harness-build.json +1 -1
- package/harness/bundles/general/codebuddy/harness-archive/SKILL.md +24 -3
- package/harness/bundles/general/codebuddy/harness-codebase-map/SKILL.md +13 -3
- package/harness/bundles/general/codebuddy/harness-knowledge-ingest/SKILL.md +13 -3
- package/harness/bundles/general/codebuddy/harness-knowledge-query/SKILL.md +13 -3
- package/harness/bundles/general/codebuddy/harness-plan/SKILL.md +13 -3
- package/harness/bundles/general/codebuddy/harness-pull/SKILL.md +1 -1
- package/harness/bundles/general/codebuddy/harness-push/SKILL.md +15 -1
- package/harness/bundles/general/codebuddy/harness-review/SKILL.md +13 -3
- package/harness/bundles/general/codebuddy/harness-run/SKILL.md +13 -3
- package/harness/bundles/general/codebuddy/harness-submit/SKILL.md +13 -3
- package/harness/bundles/general/codebuddy/harness-sync/SKILL.md +1 -1
- package/harness/bundles/general/codebuddy/harness-test/SKILL.md +13 -3
- package/harness/bundles/general/codebuddy/scripts/harness_archive.py +158 -44
- package/harness/bundles/general/codebuddy/scripts/harness_context.py +77 -3
- package/harness/bundles/general/codebuddy/scripts/harness_events.py +35 -2
- package/harness/bundles/general/codebuddy/scripts/harness_plan_finalize.py +6 -1
- package/harness/bundles/general/codex/.harness-build.json +1 -1
- package/harness/bundles/general/codex/harness-archive/SKILL.md +24 -3
- package/harness/bundles/general/codex/harness-codebase-map/SKILL.md +13 -3
- package/harness/bundles/general/codex/harness-knowledge-ingest/SKILL.md +13 -3
- package/harness/bundles/general/codex/harness-knowledge-query/SKILL.md +13 -3
- package/harness/bundles/general/codex/harness-plan/SKILL.md +13 -3
- package/harness/bundles/general/codex/harness-pull/SKILL.md +1 -1
- package/harness/bundles/general/codex/harness-push/SKILL.md +15 -1
- package/harness/bundles/general/codex/harness-review/SKILL.md +13 -3
- package/harness/bundles/general/codex/harness-run/SKILL.md +13 -3
- package/harness/bundles/general/codex/harness-submit/SKILL.md +13 -3
- package/harness/bundles/general/codex/harness-sync/SKILL.md +1 -1
- package/harness/bundles/general/codex/harness-test/SKILL.md +13 -3
- package/harness/bundles/general/codex/scripts/harness_archive.py +158 -44
- package/harness/bundles/general/codex/scripts/harness_context.py +77 -3
- package/harness/bundles/general/codex/scripts/harness_events.py +35 -2
- package/harness/bundles/general/codex/scripts/harness_plan_finalize.py +6 -1
- package/harness/bundles/general/cursor/.harness-build.json +1 -1
- package/harness/bundles/general/cursor/harness-archive/SKILL.md +24 -3
- package/harness/bundles/general/cursor/harness-codebase-map/SKILL.md +13 -3
- package/harness/bundles/general/cursor/harness-knowledge-ingest/SKILL.md +13 -3
- package/harness/bundles/general/cursor/harness-knowledge-query/SKILL.md +13 -3
- package/harness/bundles/general/cursor/harness-plan/SKILL.md +13 -3
- package/harness/bundles/general/cursor/harness-pull/SKILL.md +1 -1
- package/harness/bundles/general/cursor/harness-push/SKILL.md +15 -1
- package/harness/bundles/general/cursor/harness-review/SKILL.md +13 -3
- package/harness/bundles/general/cursor/harness-run/SKILL.md +13 -3
- package/harness/bundles/general/cursor/harness-submit/SKILL.md +13 -3
- package/harness/bundles/general/cursor/harness-sync/SKILL.md +1 -1
- package/harness/bundles/general/cursor/harness-test/SKILL.md +13 -3
- package/harness/bundles/general/cursor/scripts/harness_archive.py +158 -44
- package/harness/bundles/general/cursor/scripts/harness_context.py +77 -3
- package/harness/bundles/general/cursor/scripts/harness_events.py +35 -2
- package/harness/bundles/general/cursor/scripts/harness_plan_finalize.py +6 -1
- package/harness/bundles/java/claude-code/.harness-build.json +1 -1
- package/harness/bundles/java/claude-code/harness-apidoc/SKILL.md +13 -3
- package/harness/bundles/java/claude-code/harness-archive/SKILL.md +24 -3
- package/harness/bundles/java/claude-code/harness-codebase-map/SKILL.md +13 -3
- package/harness/bundles/java/claude-code/harness-knowledge-ingest/SKILL.md +13 -3
- package/harness/bundles/java/claude-code/harness-knowledge-query/SKILL.md +13 -3
- package/harness/bundles/java/claude-code/harness-package/SKILL.md +13 -3
- package/harness/bundles/java/claude-code/harness-plan/SKILL.md +13 -3
- package/harness/bundles/java/claude-code/harness-pull/SKILL.md +1 -1
- package/harness/bundles/java/claude-code/harness-push/SKILL.md +15 -1
- package/harness/bundles/java/claude-code/harness-review/SKILL.md +13 -3
- package/harness/bundles/java/claude-code/harness-run/SKILL.md +13 -3
- package/harness/bundles/java/claude-code/harness-submit/SKILL.md +13 -3
- package/harness/bundles/java/claude-code/harness-sync/SKILL.md +1 -1
- package/harness/bundles/java/claude-code/harness-test/SKILL.md +13 -3
- package/harness/bundles/java/claude-code/scripts/harness_archive.py +158 -44
- package/harness/bundles/java/claude-code/scripts/harness_context.py +77 -3
- package/harness/bundles/java/claude-code/scripts/harness_events.py +35 -2
- package/harness/bundles/java/claude-code/scripts/harness_plan_finalize.py +6 -1
- package/harness/bundles/java/codebuddy/.harness-build.json +1 -1
- package/harness/bundles/java/codebuddy/harness-apidoc/SKILL.md +13 -3
- package/harness/bundles/java/codebuddy/harness-archive/SKILL.md +24 -3
- package/harness/bundles/java/codebuddy/harness-codebase-map/SKILL.md +13 -3
- package/harness/bundles/java/codebuddy/harness-knowledge-ingest/SKILL.md +13 -3
- package/harness/bundles/java/codebuddy/harness-knowledge-query/SKILL.md +13 -3
- package/harness/bundles/java/codebuddy/harness-package/SKILL.md +13 -3
- package/harness/bundles/java/codebuddy/harness-plan/SKILL.md +13 -3
- package/harness/bundles/java/codebuddy/harness-pull/SKILL.md +1 -1
- package/harness/bundles/java/codebuddy/harness-push/SKILL.md +15 -1
- package/harness/bundles/java/codebuddy/harness-review/SKILL.md +13 -3
- package/harness/bundles/java/codebuddy/harness-run/SKILL.md +13 -3
- package/harness/bundles/java/codebuddy/harness-submit/SKILL.md +13 -3
- package/harness/bundles/java/codebuddy/harness-sync/SKILL.md +1 -1
- package/harness/bundles/java/codebuddy/harness-test/SKILL.md +13 -3
- package/harness/bundles/java/codebuddy/scripts/harness_archive.py +158 -44
- package/harness/bundles/java/codebuddy/scripts/harness_context.py +77 -3
- package/harness/bundles/java/codebuddy/scripts/harness_events.py +35 -2
- package/harness/bundles/java/codebuddy/scripts/harness_plan_finalize.py +6 -1
- package/harness/bundles/java/codex/.harness-build.json +1 -1
- package/harness/bundles/java/codex/harness-apidoc/SKILL.md +13 -3
- package/harness/bundles/java/codex/harness-archive/SKILL.md +24 -3
- package/harness/bundles/java/codex/harness-codebase-map/SKILL.md +13 -3
- package/harness/bundles/java/codex/harness-knowledge-ingest/SKILL.md +13 -3
- package/harness/bundles/java/codex/harness-knowledge-query/SKILL.md +13 -3
- package/harness/bundles/java/codex/harness-package/SKILL.md +13 -3
- package/harness/bundles/java/codex/harness-plan/SKILL.md +13 -3
- package/harness/bundles/java/codex/harness-pull/SKILL.md +1 -1
- package/harness/bundles/java/codex/harness-push/SKILL.md +15 -1
- package/harness/bundles/java/codex/harness-review/SKILL.md +13 -3
- package/harness/bundles/java/codex/harness-run/SKILL.md +13 -3
- package/harness/bundles/java/codex/harness-submit/SKILL.md +13 -3
- package/harness/bundles/java/codex/harness-sync/SKILL.md +1 -1
- package/harness/bundles/java/codex/harness-test/SKILL.md +13 -3
- package/harness/bundles/java/codex/scripts/harness_archive.py +158 -44
- package/harness/bundles/java/codex/scripts/harness_context.py +77 -3
- package/harness/bundles/java/codex/scripts/harness_events.py +35 -2
- package/harness/bundles/java/codex/scripts/harness_plan_finalize.py +6 -1
- package/harness/bundles/java/cursor/.harness-build.json +1 -1
- package/harness/bundles/java/cursor/harness-apidoc/SKILL.md +13 -3
- package/harness/bundles/java/cursor/harness-archive/SKILL.md +24 -3
- package/harness/bundles/java/cursor/harness-codebase-map/SKILL.md +13 -3
- package/harness/bundles/java/cursor/harness-knowledge-ingest/SKILL.md +13 -3
- package/harness/bundles/java/cursor/harness-knowledge-query/SKILL.md +13 -3
- package/harness/bundles/java/cursor/harness-package/SKILL.md +13 -3
- package/harness/bundles/java/cursor/harness-plan/SKILL.md +13 -3
- package/harness/bundles/java/cursor/harness-pull/SKILL.md +1 -1
- package/harness/bundles/java/cursor/harness-push/SKILL.md +15 -1
- package/harness/bundles/java/cursor/harness-review/SKILL.md +13 -3
- package/harness/bundles/java/cursor/harness-run/SKILL.md +13 -3
- package/harness/bundles/java/cursor/harness-submit/SKILL.md +13 -3
- package/harness/bundles/java/cursor/harness-sync/SKILL.md +1 -1
- package/harness/bundles/java/cursor/harness-test/SKILL.md +13 -3
- package/harness/bundles/java/cursor/scripts/harness_archive.py +158 -44
- package/harness/bundles/java/cursor/scripts/harness_context.py +77 -3
- package/harness/bundles/java/cursor/scripts/harness_events.py +35 -2
- package/harness/bundles/java/cursor/scripts/harness_plan_finalize.py +6 -1
- package/harness/manifests/general/claude-code.json +19 -19
- package/harness/manifests/general/codebuddy.json +19 -19
- package/harness/manifests/general/codex.json +19 -19
- package/harness/manifests/general/cursor.json +19 -19
- package/harness/manifests/java/claude-code.json +21 -21
- package/harness/manifests/java/codebuddy.json +21 -21
- package/harness/manifests/java/codex.json +21 -21
- package/harness/manifests/java/cursor.json +21 -21
- package/hunter-workflow-family.json +5 -5
- package/package.json +1 -1
|
@@ -1165,9 +1165,11 @@ def close_transition(
|
|
|
1165
1165
|
"holder": lease.get("owner"),
|
|
1166
1166
|
"leasePhase": lease.get("phase"),
|
|
1167
1167
|
}
|
|
1168
|
+
# Owner and phase already matched above, so nobody took this lease over: a
|
|
1169
|
+
# takeover rewrites the file and would have produced CONTEXT_LEASE_MISMATCH.
|
|
1170
|
+
# A lapsed deadline therefore proves only that the phase outlived the TTL.
|
|
1168
1171
|
expiry = _parse_time(lease.get("expiresAt"))
|
|
1169
|
-
|
|
1170
|
-
return {"ok": False, "code": "CONTEXT_LEASE_EXPIRED"}
|
|
1172
|
+
lease_lapsed = expiry is None or _now() >= expiry
|
|
1171
1173
|
|
|
1172
1174
|
previous_hash = transitions[-1].get("receiptHash") if transitions else None
|
|
1173
1175
|
attempt = (
|
|
@@ -1192,6 +1194,12 @@ def close_transition(
|
|
|
1192
1194
|
"closedAt": _now().isoformat(),
|
|
1193
1195
|
"previousReceiptHash": previous_hash,
|
|
1194
1196
|
}
|
|
1197
|
+
if lease_lapsed:
|
|
1198
|
+
# Recorded, not hidden: the phase ran past its lease without a renewal.
|
|
1199
|
+
receipt["leaseLapsed"] = {
|
|
1200
|
+
"expiresAt": lease.get("expiresAt"),
|
|
1201
|
+
"acquiredAt": lease.get("acquiredAt"),
|
|
1202
|
+
}
|
|
1195
1203
|
receipt["receiptHash"] = _payload_hash(receipt)
|
|
1196
1204
|
_append_ndjson(paths["transitions"], receipt)
|
|
1197
1205
|
paths["lease"].unlink(missing_ok=True)
|
|
@@ -1267,7 +1275,11 @@ def _begin_transition_unlocked(
|
|
|
1267
1275
|
"legacyBootstrap": True,
|
|
1268
1276
|
"message": (
|
|
1269
1277
|
"没有交接凭证。v2 计划应有 committed 的 meta/publication-journals/*.json "
|
|
1270
|
-
"作为补录证据;legacy 计划需先完成 plan 阶段的 close"
|
|
1278
|
+
"作为补录证据;legacy 计划需先完成 plan 阶段的 close。"
|
|
1279
|
+
"首次进入某阶段(还没有任何 transition 凭证)时不要用 context begin——"
|
|
1280
|
+
"改跑 `harness_gate.py begin --change-dir .harness/changes/<change> "
|
|
1281
|
+
"--phase <phase> --run-id <phase>_<uuid4hex> --note \"<触发指令>\"`,"
|
|
1282
|
+
"它会领取租约并写好 phase.start。"
|
|
1271
1283
|
),
|
|
1272
1284
|
}
|
|
1273
1285
|
receipt = transitions[-1]
|
|
@@ -1458,6 +1470,53 @@ def cancel_prepared_context(
|
|
|
1458
1470
|
}
|
|
1459
1471
|
|
|
1460
1472
|
|
|
1473
|
+
def renew_lease(
|
|
1474
|
+
project: Path,
|
|
1475
|
+
change: str,
|
|
1476
|
+
*,
|
|
1477
|
+
executor: str,
|
|
1478
|
+
ttl_seconds: int = 3600,
|
|
1479
|
+
) -> dict[str, Any]:
|
|
1480
|
+
"""Extend the current phase lease in place.
|
|
1481
|
+
|
|
1482
|
+
Only the recorded owner may renew, and only while no other executor has
|
|
1483
|
+
taken the lease over. Renewal touches nothing but ``expiresAt``, so it is
|
|
1484
|
+
safe to call on a heartbeat during a long phase.
|
|
1485
|
+
"""
|
|
1486
|
+
project = project.resolve()
|
|
1487
|
+
try:
|
|
1488
|
+
_, _, state_root = _contract(project, change)
|
|
1489
|
+
except ValueError as exc:
|
|
1490
|
+
return {"ok": False, "code": _contract_error_code(exc), "error": str(exc)}
|
|
1491
|
+
paths = _paths(state_root)
|
|
1492
|
+
if not paths["lease"].is_file():
|
|
1493
|
+
return {"ok": False, "code": "CONTEXT_LEASE_REQUIRED"}
|
|
1494
|
+
with _exclusive_state_lock(paths["runtime"] / "branch-selection.lock"):
|
|
1495
|
+
try:
|
|
1496
|
+
lease = _read_json(paths["lease"])
|
|
1497
|
+
except (OSError, ValueError, json.JSONDecodeError):
|
|
1498
|
+
return {"ok": False, "code": "CONTEXT_LEASE_INVALID"}
|
|
1499
|
+
if lease.get("owner") != executor:
|
|
1500
|
+
return {
|
|
1501
|
+
"ok": False,
|
|
1502
|
+
"code": "CONTEXT_LEASE_MISMATCH",
|
|
1503
|
+
"holder": lease.get("owner"),
|
|
1504
|
+
}
|
|
1505
|
+
expiry = _parse_time(lease.get("expiresAt"))
|
|
1506
|
+
lapsed = expiry is None or _now() >= expiry
|
|
1507
|
+
lease["expiresAt"] = (
|
|
1508
|
+
_now() + dt.timedelta(seconds=max(1, int(ttl_seconds)))
|
|
1509
|
+
).isoformat()
|
|
1510
|
+
lease["renewedAt"] = _now().isoformat()
|
|
1511
|
+
_write_json_atomic(paths["lease"], lease)
|
|
1512
|
+
return {
|
|
1513
|
+
"ok": True,
|
|
1514
|
+
"code": "CONTEXT_LEASE_RENEWED",
|
|
1515
|
+
"lease": lease,
|
|
1516
|
+
"wasLapsed": lapsed,
|
|
1517
|
+
}
|
|
1518
|
+
|
|
1519
|
+
|
|
1461
1520
|
def context_view(project: Path, change: str) -> dict[str, Any]:
|
|
1462
1521
|
project = Path(project).resolve()
|
|
1463
1522
|
try:
|
|
@@ -1536,6 +1595,14 @@ def build_parser() -> argparse.ArgumentParser:
|
|
|
1536
1595
|
begin.add_argument("--change", required=True)
|
|
1537
1596
|
begin.add_argument("--phase", required=True)
|
|
1538
1597
|
begin.add_argument("--executor", required=True)
|
|
1598
|
+
renew = sub.add_parser(
|
|
1599
|
+
"renew", help="extend the current phase lease without touching context"
|
|
1600
|
+
)
|
|
1601
|
+
renew.add_argument("--json", action="store_true")
|
|
1602
|
+
renew.add_argument("--project", required=True, type=Path)
|
|
1603
|
+
renew.add_argument("--change", required=True)
|
|
1604
|
+
renew.add_argument("--executor", required=True)
|
|
1605
|
+
renew.add_argument("--ttl-seconds", type=int, default=3600)
|
|
1539
1606
|
configure = sub.add_parser("configure-plan")
|
|
1540
1607
|
configure.add_argument("--json", action="store_true")
|
|
1541
1608
|
configure.add_argument("--project", required=True, type=Path)
|
|
@@ -1588,6 +1655,13 @@ def main(argv: list[str] | None = None) -> int:
|
|
|
1588
1655
|
phase=args.phase,
|
|
1589
1656
|
executor=args.executor,
|
|
1590
1657
|
)
|
|
1658
|
+
elif args.command == "renew":
|
|
1659
|
+
result = renew_lease(
|
|
1660
|
+
args.project,
|
|
1661
|
+
args.change,
|
|
1662
|
+
executor=args.executor,
|
|
1663
|
+
ttl_seconds=args.ttl_seconds,
|
|
1664
|
+
)
|
|
1591
1665
|
elif args.command == "configure-plan":
|
|
1592
1666
|
result = configure_phase_plan(
|
|
1593
1667
|
args.project,
|
|
@@ -1904,6 +1904,7 @@ def cmd_append(args: argparse.Namespace) -> int:
|
|
|
1904
1904
|
file=sys.stderr,
|
|
1905
1905
|
)
|
|
1906
1906
|
phase_closed = False
|
|
1907
|
+
phase_start_duplicate = False
|
|
1907
1908
|
projection_error: str | None = None
|
|
1908
1909
|
attempt_error: str | None = None
|
|
1909
1910
|
auto_sealed_events: list[dict[str, Any]] = []
|
|
@@ -1951,10 +1952,25 @@ def cmd_append(args: argparse.Namespace) -> int:
|
|
|
1951
1952
|
except ValueError as exc:
|
|
1952
1953
|
projection_error = str(exc)
|
|
1953
1954
|
elif args.type == "phase.start":
|
|
1955
|
+
# Re-appending the same (phase, run_id) is the caller restating a
|
|
1956
|
+
# start that already exists — `gate begin` writes one, and the
|
|
1957
|
+
# logging protocol used to ask for a second. Treat it as a no-op
|
|
1958
|
+
# rather than a duplicate; two identical phase.start events make
|
|
1959
|
+
# `plan finalize` unrecoverable via PHASE_START_DUPLICATE.
|
|
1960
|
+
if run_id is not None and any(
|
|
1961
|
+
item.get("phase") == args.phase
|
|
1962
|
+
and item.get("type") == "phase.start"
|
|
1963
|
+
and item.get("run_id") == run_id
|
|
1964
|
+
for item in existing_events
|
|
1965
|
+
):
|
|
1966
|
+
phase_start_duplicate = True
|
|
1954
1967
|
# HH-WF-20260730-001: a new phase.start must not be appended
|
|
1955
1968
|
# while a prior attempt for the same phase is still open —
|
|
1956
1969
|
# auto-seal it first so timing never has two open attempts.
|
|
1957
|
-
open_attempts =
|
|
1970
|
+
open_attempts = (
|
|
1971
|
+
[] if phase_start_duplicate
|
|
1972
|
+
else open_attempts_for_phase(existing_events, args.phase)
|
|
1973
|
+
)
|
|
1958
1974
|
if open_attempts:
|
|
1959
1975
|
inferred_reason = infer_auto_seal_reason(
|
|
1960
1976
|
open_attempts[-1].get("events") or []
|
|
@@ -1964,7 +1980,12 @@ def cmd_append(args: argparse.Namespace) -> int:
|
|
|
1964
1980
|
phase=args.phase,
|
|
1965
1981
|
seal_reason=inferred_reason,
|
|
1966
1982
|
)
|
|
1967
|
-
if
|
|
1983
|
+
if (
|
|
1984
|
+
not phase_closed
|
|
1985
|
+
and not phase_start_duplicate
|
|
1986
|
+
and projection_error is None
|
|
1987
|
+
and attempt_error is None
|
|
1988
|
+
):
|
|
1968
1989
|
for seal_event in auto_sealed_events:
|
|
1969
1990
|
atomic_append_line(
|
|
1970
1991
|
path,
|
|
@@ -1976,6 +1997,18 @@ def cmd_append(args: argparse.Namespace) -> int:
|
|
|
1976
1997
|
)
|
|
1977
1998
|
except (OSError, TimeoutError, ValueError) as exc:
|
|
1978
1999
|
return emit_error(f"append failed: {exc}", as_json=as_json)
|
|
2000
|
+
if phase_start_duplicate:
|
|
2001
|
+
if as_json:
|
|
2002
|
+
print(json.dumps({
|
|
2003
|
+
"ok": True,
|
|
2004
|
+
"skipped": True,
|
|
2005
|
+
"reason": "phase-start-already-recorded",
|
|
2006
|
+
"phase": args.phase,
|
|
2007
|
+
"run_id": event.get("run_id"),
|
|
2008
|
+
}, ensure_ascii=False))
|
|
2009
|
+
else:
|
|
2010
|
+
print("ok (phase.start already recorded for this run id)")
|
|
2011
|
+
return 0
|
|
1979
2012
|
if phase_closed:
|
|
1980
2013
|
return emit_error(
|
|
1981
2014
|
"PHASE_ALREADY_CLOSED: refusing a second phase.end for the same "
|
|
@@ -735,7 +735,12 @@ def _validate_plan_start(
|
|
|
735
735
|
if len(matching) > 1:
|
|
736
736
|
return _result_error(
|
|
737
737
|
"PHASE_START_DUPLICATE",
|
|
738
|
-
f"found {len(matching)} matching plan phase.start events"
|
|
738
|
+
f"found {len(matching)} matching plan phase.start events for run id "
|
|
739
|
+
f"{run_id}. `harness_gate.py begin` already appends phase.start; a "
|
|
740
|
+
"second manual append of the same run id creates this state. Recover "
|
|
741
|
+
"with a fresh run id and the next attempt number (attempts are global "
|
|
742
|
+
"per phase, not per run id): append phase.start with both, then "
|
|
743
|
+
"finalize with the same pair.",
|
|
739
744
|
)
|
|
740
745
|
return {"ok": True, "phaseStartCount": 1}
|
|
741
746
|
|
|
@@ -3,7 +3,7 @@ name: harness-archive
|
|
|
3
3
|
description: 归档所有变更产出(计划/测试报告/审查/SQL/API文档)到 .harness/archive/,含归档元数据。仅当用户显式调用
|
|
4
4
|
/harness-archive 时使用;不得被其他阶段 skill 自动接续触发。
|
|
5
5
|
---
|
|
6
|
-
<!-- generated by harness_deploy.py; core=
|
|
6
|
+
<!-- generated by harness_deploy.py; core=7fdcf0b62789ff65; overlay=none; agent=codex; do not edit -->
|
|
7
7
|
# harness-archive — 变更归档
|
|
8
8
|
|
|
9
9
|
## Purpose
|
|
@@ -144,10 +144,21 @@ python <skills-root>/scripts/harness_archive.py republish --change <change-key>
|
|
|
144
144
|
python <skills-root>/scripts/harness_archive.py republish --change <change-key> --json # 重建并上传
|
|
145
145
|
```
|
|
146
146
|
|
|
147
|
+
脚本在 `<skills-root>/scripts/` 共享,**不在 `harness-archive/scripts/`**(见 read-protocol 第 0 条)。
|
|
148
|
+
|
|
149
|
+
在项目根内任意目录直接跑即可,**不需要** `--project`。
|
|
150
|
+
|
|
147
151
|
归档目录是封存的(after-manifest 覆盖其每个字节),因此 `republish` **不写回**归档目录:
|
|
148
152
|
缺失的 `candidates/knowledge.json` 只在内存中按已归档 summary 生成并放进包里。平台的知识
|
|
149
153
|
条目只来自这个包;`--scope all` 走的是 branch_file 通道,不进知识管道。
|
|
150
154
|
|
|
155
|
+
**服务端一个 change key 只存一个不可变包**,所以补传只适用于「从未成功上传」与「字节完全一致的
|
|
156
|
+
重试」。已 durable 的 change 会在本地就判出结果:字节一致 → `ARCHIVE_ALREADY_PUBLISHED`
|
|
157
|
+
(exit 0,不重传);字节不同 → `ARCHIVE_REMOTE_IMMUTABLE_CONFLICT`(exit 1,**不发起上传**)。
|
|
158
|
+
**已发布归档无法从客户端补知识条目**——注入候选必然改变字节,服务端拒绝替换;要补需要平台侧
|
|
159
|
+
提供重新索引或归档版本化能力。
|
|
160
|
+
只想重试**同一个包**(例如知识索引失败)时加 `--no-knowledge-injection`,它按封存目录原样重建,字节与已存包一致。
|
|
161
|
+
|
|
151
162
|
### 三、文件移动只用内置工具或 PowerShell
|
|
152
163
|
|
|
153
164
|
移动用 Read+Write+验证 或 PowerShell;**禁止 Bash mv/cp/rm**。移动失败时不删除原目录,报错退出让用户手动处理。
|
|
@@ -247,13 +258,23 @@ git 命令通过 `powershell.exe -Command "..."` 执行;archive-meta.md 和 su
|
|
|
247
258
|
|
|
248
259
|
`events.ndjson` 为唯一事实源(schema_version 3,兼容读取 v1/v2;`note` 承载人类可读摘要);`logs/execution-log.md` 由 `harness_events.py` 渲染,**禁止用 Write/Edit 直接维护**。直接修改的内容会在 `phase.end` 或 finalize 时被完整重建覆盖,属于数据丢失;需要保留的详情必须进入事件 `note`。结构 → [[../protocols/report-pipeline-protocol.md|report-pipeline-protocol]]
|
|
249
260
|
|
|
261
|
+
**`phase.start` 由 `harness_gate.py begin` 写,不要再手工追加一次。** 两条同 `run-id` 的
|
|
262
|
+
`phase.start` 会让 `plan finalize` 以 `PHASE_START_DUPLICATE` 卡死,而且手工那次会先触发
|
|
263
|
+
auto-seal、把正在开始的 attempt 封成 `RECOVERED`。要补触发指令说明就带 `--note` 跑 `gate begin`。
|
|
264
|
+
(重复追加现已按 `(phase, run-id)` 判为幂等 no-op,但依赖它不如不写。)
|
|
265
|
+
|
|
250
266
|
```powershell
|
|
251
|
-
|
|
267
|
+
# 阶段开始:gate begin 负责,note 在这里给
|
|
268
|
+
python <skills-root>/scripts/harness_gate.py begin --change-dir ".harness/changes/<change-name>" --phase <phase> --note "<触发指令>"
|
|
269
|
+
# 阶段中的其他事件才用 append
|
|
270
|
+
python <skills-root>/scripts/harness_events.py append --change-dir ".harness/changes/<change-name>" --phase <phase> --type <command|issue|verification> --run-id <phase-run-id> --note "<摘要>"
|
|
252
271
|
```
|
|
253
272
|
|
|
254
273
|
> **脚本接线**:`harness_events.py append`;`harness_archive.py finalize`;`harness_preflight.py check`;`harness_ledger.py can-reuse`;`harness_service.py ensure/stop`(须 `--files`/`serviceStart.inputFiles`)。JSON 输出按 D13 护栏解读。
|
|
255
274
|
|
|
256
275
|
> **Task 4 §6.1 写入契约**:普通 `append` = 加锁 -> 追加一行 -> fsync -> 解锁,**不 load 历史、不渲染**(O(1),跨进程锁 `events.ndjson.lock`,UUID 用完整 `uuid4().hex` 无需去重扫描)。仅 `--type phase.end` append 在追加成功后渲染一次 `execution-log.md`;显式 `harness_events.py render` 随时从完整 events 重建;`harness_archive.py finalize` 在 collect 前强制 render 一次。高频 command append 期间 log 可能滞后,phase 边界保持最新。
|
|
257
276
|
|
|
258
|
-
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n
|
|
277
|
+
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n>`,不得覆盖或伪装成一次执行。**`attempt` 按 phase 全局递增,不是按 run-id**:一个 run-id 只绑定一个 attempt,重试必须「新 run-id + 下一个 attempt」两者同时换,只换其一会撞 `EVENT_ATTEMPT_CONFLICT` 或 `PHASE_ALREADY_CLOSED`。已发布的 plan 用 `harness_plan_finalize.py republish` 自动分配两者。
|
|
278
|
+
|
|
279
|
+
阶段跑得久(plan/run 常见)时用 `harness_context.py renew --project . --change <cn> --executor <tool>` 续租;租约到期本身不再阻断 `close`(同一 owner 的过期租约不构成冲突,收据里记 `leaseLapsed`),但续租能让 `view` 的状态如实反映在跑。跨工具继续执行时写 `--executor-tool <codex|claude-code|codebuddy|cursor>`,并在接棒事件写 `--handoff-from-tool` / `--handoff-reason`;也可由 `HUNTER_HARNESS_TOOL/AGENT/MODEL/RUN_ID` 环境变量统一注入。
|
|
259
280
|
> 片段:logging · phase=`archive`
|
|
@@ -3,7 +3,7 @@ name: harness-codebase-map
|
|
|
3
3
|
description: 分析现有代码库并生成 .harness/codebase/map/
|
|
4
4
|
下的结构化代码库地图。用于陌生项目理解、重大改造前梳理、harness-sync 提示 map 过期后的人工刷新。
|
|
5
5
|
---
|
|
6
|
-
<!-- generated by harness_deploy.py; core=
|
|
6
|
+
<!-- generated by harness_deploy.py; core=7fdcf0b62789ff65; overlay=none; agent=codex; do not edit -->
|
|
7
7
|
# harness-codebase-map — 代码库地图
|
|
8
8
|
|
|
9
9
|
## Purpose
|
|
@@ -94,13 +94,23 @@ description: 分析现有代码库并生成 .harness/codebase/map/
|
|
|
94
94
|
|
|
95
95
|
`events.ndjson` 为唯一事实源(schema_version 3,兼容读取 v1/v2;`note` 承载人类可读摘要);`logs/execution-log.md` 由 `harness_events.py` 渲染,**禁止用 Write/Edit 直接维护**。直接修改的内容会在 `phase.end` 或 finalize 时被完整重建覆盖,属于数据丢失;需要保留的详情必须进入事件 `note`。结构 → [[../protocols/report-pipeline-protocol.md|report-pipeline-protocol]]
|
|
96
96
|
|
|
97
|
+
**`phase.start` 由 `harness_gate.py begin` 写,不要再手工追加一次。** 两条同 `run-id` 的
|
|
98
|
+
`phase.start` 会让 `plan finalize` 以 `PHASE_START_DUPLICATE` 卡死,而且手工那次会先触发
|
|
99
|
+
auto-seal、把正在开始的 attempt 封成 `RECOVERED`。要补触发指令说明就带 `--note` 跑 `gate begin`。
|
|
100
|
+
(重复追加现已按 `(phase, run-id)` 判为幂等 no-op,但依赖它不如不写。)
|
|
101
|
+
|
|
97
102
|
```powershell
|
|
98
|
-
|
|
103
|
+
# 阶段开始:gate begin 负责,note 在这里给
|
|
104
|
+
python <skills-root>/scripts/harness_gate.py begin --change-dir ".harness/changes/<change-name>" --phase <phase> --note "<触发指令>"
|
|
105
|
+
# 阶段中的其他事件才用 append
|
|
106
|
+
python <skills-root>/scripts/harness_events.py append --change-dir ".harness/changes/<change-name>" --phase <phase> --type <command|issue|verification> --run-id <phase-run-id> --note "<摘要>"
|
|
99
107
|
```
|
|
100
108
|
|
|
101
109
|
> **脚本接线**:`harness_events.py append`;`harness_archive.py finalize`;`harness_preflight.py check`;`harness_ledger.py can-reuse`;`harness_service.py ensure/stop`(须 `--files`/`serviceStart.inputFiles`)。JSON 输出按 D13 护栏解读。
|
|
102
110
|
|
|
103
111
|
> **Task 4 §6.1 写入契约**:普通 `append` = 加锁 -> 追加一行 -> fsync -> 解锁,**不 load 历史、不渲染**(O(1),跨进程锁 `events.ndjson.lock`,UUID 用完整 `uuid4().hex` 无需去重扫描)。仅 `--type phase.end` append 在追加成功后渲染一次 `execution-log.md`;显式 `harness_events.py render` 随时从完整 events 重建;`harness_archive.py finalize` 在 collect 前强制 render 一次。高频 command append 期间 log 可能滞后,phase 边界保持最新。
|
|
104
112
|
|
|
105
|
-
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n
|
|
113
|
+
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n>`,不得覆盖或伪装成一次执行。**`attempt` 按 phase 全局递增,不是按 run-id**:一个 run-id 只绑定一个 attempt,重试必须「新 run-id + 下一个 attempt」两者同时换,只换其一会撞 `EVENT_ATTEMPT_CONFLICT` 或 `PHASE_ALREADY_CLOSED`。已发布的 plan 用 `harness_plan_finalize.py republish` 自动分配两者。
|
|
114
|
+
|
|
115
|
+
阶段跑得久(plan/run 常见)时用 `harness_context.py renew --project . --change <cn> --executor <tool>` 续租;租约到期本身不再阻断 `close`(同一 owner 的过期租约不构成冲突,收据里记 `leaseLapsed`),但续租能让 `view` 的状态如实反映在跑。跨工具继续执行时写 `--executor-tool <codex|claude-code|codebuddy|cursor>`,并在接棒事件写 `--handoff-from-tool` / `--handoff-reason`;也可由 `HUNTER_HARNESS_TOOL/AGENT/MODEL/RUN_ID` 环境变量统一注入。
|
|
106
116
|
> 片段:logging · phase=`codebase-map`;默认写 reports,有未归档变更时另写 change events
|
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
name: harness-knowledge-ingest
|
|
3
3
|
description: 确认归档 ZIP 已上传并由 Hunter Platform 在服务端解包、校验和入库。客户端不再构建或维护本地知识索引。
|
|
4
4
|
---
|
|
5
|
-
<!-- generated by harness_deploy.py; core=
|
|
5
|
+
<!-- generated by harness_deploy.py; core=7fdcf0b62789ff65; overlay=none; agent=codex; do not edit -->
|
|
6
6
|
# harness-knowledge-ingest
|
|
7
7
|
|
|
8
8
|
知识 ingest 完全由 Hunter Platform 负责。归档完成时,客户端生成一个确定性 ZIP;
|
|
@@ -81,13 +81,23 @@ ZIP 已按收据策略清理,且平台下载接口可以恢复原包。
|
|
|
81
81
|
|
|
82
82
|
`events.ndjson` 为唯一事实源(schema_version 3,兼容读取 v1/v2;`note` 承载人类可读摘要);`logs/execution-log.md` 由 `harness_events.py` 渲染,**禁止用 Write/Edit 直接维护**。直接修改的内容会在 `phase.end` 或 finalize 时被完整重建覆盖,属于数据丢失;需要保留的详情必须进入事件 `note`。结构 → [[../protocols/report-pipeline-protocol.md|report-pipeline-protocol]]
|
|
83
83
|
|
|
84
|
+
**`phase.start` 由 `harness_gate.py begin` 写,不要再手工追加一次。** 两条同 `run-id` 的
|
|
85
|
+
`phase.start` 会让 `plan finalize` 以 `PHASE_START_DUPLICATE` 卡死,而且手工那次会先触发
|
|
86
|
+
auto-seal、把正在开始的 attempt 封成 `RECOVERED`。要补触发指令说明就带 `--note` 跑 `gate begin`。
|
|
87
|
+
(重复追加现已按 `(phase, run-id)` 判为幂等 no-op,但依赖它不如不写。)
|
|
88
|
+
|
|
84
89
|
```powershell
|
|
85
|
-
|
|
90
|
+
# 阶段开始:gate begin 负责,note 在这里给
|
|
91
|
+
python <skills-root>/scripts/harness_gate.py begin --change-dir ".harness/changes/<change-name>" --phase <phase> --note "<触发指令>"
|
|
92
|
+
# 阶段中的其他事件才用 append
|
|
93
|
+
python <skills-root>/scripts/harness_events.py append --change-dir ".harness/changes/<change-name>" --phase <phase> --type <command|issue|verification> --run-id <phase-run-id> --note "<摘要>"
|
|
86
94
|
```
|
|
87
95
|
|
|
88
96
|
> **脚本接线**:`harness_events.py append`;`harness_archive.py finalize`;`harness_preflight.py check`;`harness_ledger.py can-reuse`;`harness_service.py ensure/stop`(须 `--files`/`serviceStart.inputFiles`)。JSON 输出按 D13 护栏解读。
|
|
89
97
|
|
|
90
98
|
> **Task 4 §6.1 写入契约**:普通 `append` = 加锁 -> 追加一行 -> fsync -> 解锁,**不 load 历史、不渲染**(O(1),跨进程锁 `events.ndjson.lock`,UUID 用完整 `uuid4().hex` 无需去重扫描)。仅 `--type phase.end` append 在追加成功后渲染一次 `execution-log.md`;显式 `harness_events.py render` 随时从完整 events 重建;`harness_archive.py finalize` 在 collect 前强制 render 一次。高频 command append 期间 log 可能滞后,phase 边界保持最新。
|
|
91
99
|
|
|
92
|
-
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n
|
|
100
|
+
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n>`,不得覆盖或伪装成一次执行。**`attempt` 按 phase 全局递增,不是按 run-id**:一个 run-id 只绑定一个 attempt,重试必须「新 run-id + 下一个 attempt」两者同时换,只换其一会撞 `EVENT_ATTEMPT_CONFLICT` 或 `PHASE_ALREADY_CLOSED`。已发布的 plan 用 `harness_plan_finalize.py republish` 自动分配两者。
|
|
101
|
+
|
|
102
|
+
阶段跑得久(plan/run 常见)时用 `harness_context.py renew --project . --change <cn> --executor <tool>` 续租;租约到期本身不再阻断 `close`(同一 owner 的过期租约不构成冲突,收据里记 `leaseLapsed`),但续租能让 `view` 的状态如实反映在跑。跨工具继续执行时写 `--executor-tool <codex|claude-code|codebuddy|cursor>`,并在接棒事件写 `--handoff-from-tool` / `--handoff-reason`;也可由 `HUNTER_HARNESS_TOOL/AGENT/MODEL/RUN_ID` 环境变量统一注入。
|
|
93
103
|
> 片段:logging · phase=`knowledge-ingest` · 记录 package hash、服务端状态与失败重试路径
|
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
name: harness-knowledge-query
|
|
3
3
|
description: 在规划、实现或排查前,通过 hunter-harness CLI 查询远端项目知识库。远端不可用时直接报告不可用,不建立本地索引或离线回退。
|
|
4
4
|
---
|
|
5
|
-
<!-- generated by harness_deploy.py; core=
|
|
5
|
+
<!-- generated by harness_deploy.py; core=7fdcf0b62789ff65; overlay=none; agent=codex; do not edit -->
|
|
6
6
|
# harness-knowledge-query
|
|
7
7
|
|
|
8
8
|
项目知识以 Hunter Platform 的服务端索引为唯一真源。客户端只提交查询并消费结果:
|
|
@@ -75,13 +75,23 @@ powershell.exe -Command "npx hunter-harness knowledge query '<用户需求原文
|
|
|
75
75
|
|
|
76
76
|
`events.ndjson` 为唯一事实源(schema_version 3,兼容读取 v1/v2;`note` 承载人类可读摘要);`logs/execution-log.md` 由 `harness_events.py` 渲染,**禁止用 Write/Edit 直接维护**。直接修改的内容会在 `phase.end` 或 finalize 时被完整重建覆盖,属于数据丢失;需要保留的详情必须进入事件 `note`。结构 → [[../protocols/report-pipeline-protocol.md|report-pipeline-protocol]]
|
|
77
77
|
|
|
78
|
+
**`phase.start` 由 `harness_gate.py begin` 写,不要再手工追加一次。** 两条同 `run-id` 的
|
|
79
|
+
`phase.start` 会让 `plan finalize` 以 `PHASE_START_DUPLICATE` 卡死,而且手工那次会先触发
|
|
80
|
+
auto-seal、把正在开始的 attempt 封成 `RECOVERED`。要补触发指令说明就带 `--note` 跑 `gate begin`。
|
|
81
|
+
(重复追加现已按 `(phase, run-id)` 判为幂等 no-op,但依赖它不如不写。)
|
|
82
|
+
|
|
78
83
|
```powershell
|
|
79
|
-
|
|
84
|
+
# 阶段开始:gate begin 负责,note 在这里给
|
|
85
|
+
python <skills-root>/scripts/harness_gate.py begin --change-dir ".harness/changes/<change-name>" --phase <phase> --note "<触发指令>"
|
|
86
|
+
# 阶段中的其他事件才用 append
|
|
87
|
+
python <skills-root>/scripts/harness_events.py append --change-dir ".harness/changes/<change-name>" --phase <phase> --type <command|issue|verification> --run-id <phase-run-id> --note "<摘要>"
|
|
80
88
|
```
|
|
81
89
|
|
|
82
90
|
> **脚本接线**:`harness_events.py append`;`harness_archive.py finalize`;`harness_preflight.py check`;`harness_ledger.py can-reuse`;`harness_service.py ensure/stop`(须 `--files`/`serviceStart.inputFiles`)。JSON 输出按 D13 护栏解读。
|
|
83
91
|
|
|
84
92
|
> **Task 4 §6.1 写入契约**:普通 `append` = 加锁 -> 追加一行 -> fsync -> 解锁,**不 load 历史、不渲染**(O(1),跨进程锁 `events.ndjson.lock`,UUID 用完整 `uuid4().hex` 无需去重扫描)。仅 `--type phase.end` append 在追加成功后渲染一次 `execution-log.md`;显式 `harness_events.py render` 随时从完整 events 重建;`harness_archive.py finalize` 在 collect 前强制 render 一次。高频 command append 期间 log 可能滞后,phase 边界保持最新。
|
|
85
93
|
|
|
86
|
-
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n
|
|
94
|
+
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n>`,不得覆盖或伪装成一次执行。**`attempt` 按 phase 全局递增,不是按 run-id**:一个 run-id 只绑定一个 attempt,重试必须「新 run-id + 下一个 attempt」两者同时换,只换其一会撞 `EVENT_ATTEMPT_CONFLICT` 或 `PHASE_ALREADY_CLOSED`。已发布的 plan 用 `harness_plan_finalize.py republish` 自动分配两者。
|
|
95
|
+
|
|
96
|
+
阶段跑得久(plan/run 常见)时用 `harness_context.py renew --project . --change <cn> --executor <tool>` 续租;租约到期本身不再阻断 `close`(同一 owner 的过期租约不构成冲突,收据里记 `leaseLapsed`),但续租能让 `view` 的状态如实反映在跑。跨工具继续执行时写 `--executor-tool <codex|claude-code|codebuddy|cursor>`,并在接棒事件写 `--handoff-from-tool` / `--handoff-reason`;也可由 `HUNTER_HARNESS_TOOL/AGENT/MODEL/RUN_ID` 环境变量统一注入。
|
|
87
97
|
> 片段:logging · phase=`knowledge-query` · 成功记录命中摘要,失败记录远端错误码且不做本地回退
|
|
@@ -3,7 +3,7 @@ name: harness-plan
|
|
|
3
3
|
description: 将需求转化为设计文档+实施计划+完整测试场景表,必须在编码前完成。仅当用户显式调用 /harness-plan 或明确要求进入
|
|
4
4
|
Harness 规划阶段时使用;不得因用户描述需求就自动触发。
|
|
5
5
|
---
|
|
6
|
-
<!-- generated by harness_deploy.py; core=
|
|
6
|
+
<!-- generated by harness_deploy.py; core=7fdcf0b62789ff65; overlay=none; agent=codex; do not edit -->
|
|
7
7
|
# harness-plan — 需求规划
|
|
8
8
|
|
|
9
9
|
## Purpose
|
|
@@ -134,13 +134,23 @@ change-name 范围变更 → 提示重命名或记 🟡WARN(→ `reference.md`
|
|
|
134
134
|
|
|
135
135
|
`events.ndjson` 为唯一事实源(schema_version 3,兼容读取 v1/v2;`note` 承载人类可读摘要);`logs/execution-log.md` 由 `harness_events.py` 渲染,**禁止用 Write/Edit 直接维护**。直接修改的内容会在 `phase.end` 或 finalize 时被完整重建覆盖,属于数据丢失;需要保留的详情必须进入事件 `note`。结构 → [[../protocols/report-pipeline-protocol.md|report-pipeline-protocol]]
|
|
136
136
|
|
|
137
|
+
**`phase.start` 由 `harness_gate.py begin` 写,不要再手工追加一次。** 两条同 `run-id` 的
|
|
138
|
+
`phase.start` 会让 `plan finalize` 以 `PHASE_START_DUPLICATE` 卡死,而且手工那次会先触发
|
|
139
|
+
auto-seal、把正在开始的 attempt 封成 `RECOVERED`。要补触发指令说明就带 `--note` 跑 `gate begin`。
|
|
140
|
+
(重复追加现已按 `(phase, run-id)` 判为幂等 no-op,但依赖它不如不写。)
|
|
141
|
+
|
|
137
142
|
```powershell
|
|
138
|
-
|
|
143
|
+
# 阶段开始:gate begin 负责,note 在这里给
|
|
144
|
+
python <skills-root>/scripts/harness_gate.py begin --change-dir ".harness/changes/<change-name>" --phase <phase> --note "<触发指令>"
|
|
145
|
+
# 阶段中的其他事件才用 append
|
|
146
|
+
python <skills-root>/scripts/harness_events.py append --change-dir ".harness/changes/<change-name>" --phase <phase> --type <command|issue|verification> --run-id <phase-run-id> --note "<摘要>"
|
|
139
147
|
```
|
|
140
148
|
|
|
141
149
|
> **脚本接线**:`harness_events.py append`;`harness_archive.py finalize`;`harness_preflight.py check`;`harness_ledger.py can-reuse`;`harness_service.py ensure/stop`(须 `--files`/`serviceStart.inputFiles`)。JSON 输出按 D13 护栏解读。
|
|
142
150
|
|
|
143
151
|
> **Task 4 §6.1 写入契约**:普通 `append` = 加锁 -> 追加一行 -> fsync -> 解锁,**不 load 历史、不渲染**(O(1),跨进程锁 `events.ndjson.lock`,UUID 用完整 `uuid4().hex` 无需去重扫描)。仅 `--type phase.end` append 在追加成功后渲染一次 `execution-log.md`;显式 `harness_events.py render` 随时从完整 events 重建;`harness_archive.py finalize` 在 collect 前强制 render 一次。高频 command append 期间 log 可能滞后,phase 边界保持最新。
|
|
144
152
|
|
|
145
|
-
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n
|
|
153
|
+
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n>`,不得覆盖或伪装成一次执行。**`attempt` 按 phase 全局递增,不是按 run-id**:一个 run-id 只绑定一个 attempt,重试必须「新 run-id + 下一个 attempt」两者同时换,只换其一会撞 `EVENT_ATTEMPT_CONFLICT` 或 `PHASE_ALREADY_CLOSED`。已发布的 plan 用 `harness_plan_finalize.py republish` 自动分配两者。
|
|
154
|
+
|
|
155
|
+
阶段跑得久(plan/run 常见)时用 `harness_context.py renew --project . --change <cn> --executor <tool>` 续租;租约到期本身不再阻断 `close`(同一 owner 的过期租约不构成冲突,收据里记 `leaseLapsed`),但续租能让 `view` 的状态如实反映在跑。跨工具继续执行时写 `--executor-tool <codex|claude-code|codebuddy|cursor>`,并在接棒事件写 `--handoff-from-tool` / `--handoff-reason`;也可由 `HUNTER_HARNESS_TOOL/AGENT/MODEL/RUN_ID` 环境变量统一注入。
|
|
146
156
|
> 片段:logging · phase=`plan`
|
|
@@ -3,7 +3,7 @@ name: harness-pull
|
|
|
3
3
|
description: 从 Hunter Platform 下拉配置/规则/架构/指令(及显式来源分支的分支文件恢复)。仅当用户显式调用
|
|
4
4
|
/harness-pull 或明确说'从平台拉取/恢复'时使用;不得自动触发。
|
|
5
5
|
---
|
|
6
|
-
<!-- generated by harness_deploy.py; core=
|
|
6
|
+
<!-- generated by harness_deploy.py; core=7fdcf0b62789ff65; overlay=none; agent=codex; do not edit -->
|
|
7
7
|
# harness-pull — 从 Hunter Platform 下拉与恢复
|
|
8
8
|
|
|
9
9
|
## Purpose
|
|
@@ -3,7 +3,7 @@ name: harness-push
|
|
|
3
3
|
description: 上传本地配置/规则/架构/指令(及显式归档)到 Hunter Platform。仅当用户显式调用 /harness-push
|
|
4
4
|
或明确说'上传到平台'时使用;不得因存在本地修改就自动触发。
|
|
5
5
|
---
|
|
6
|
-
<!-- generated by harness_deploy.py; core=
|
|
6
|
+
<!-- generated by harness_deploy.py; core=7fdcf0b62789ff65; overlay=none; agent=codex; do not edit -->
|
|
7
7
|
# harness-push — 上传到 Hunter Platform
|
|
8
8
|
|
|
9
9
|
## Purpose
|
|
@@ -39,8 +39,22 @@ npx hunter-harness harness-push --scope archive --change <change-key> --json
|
|
|
39
39
|
python <skills-root>/scripts/harness_archive.py republish --change <change-key> --json
|
|
40
40
|
```
|
|
41
41
|
|
|
42
|
+
`<skills-root>` 是已部署 skills 的根目录,实际形态是 `.codebuddy/skills/`(`.claude`、`.cursor`、
|
|
43
|
+
`.codex` 同理)。**脚本在它下面的 `scripts/` 里共享,不在 `harness-push/scripts/`** —— 已有四份
|
|
44
|
+
执行日志先猜成后者、报 `No such file` 再回头找。
|
|
45
|
+
|
|
42
46
|
`--scope archive` 没有待发布 claim 时返回 `PUSH_PULL_ARCHIVE_NO_PENDING_CLAIM`(exit 5),
|
|
43
47
|
stderr 里直接给出上面的 republish 命令;`--dry-run` 会列出本地可补传的归档目录,不取租约。
|
|
48
|
+
`republish` 在项目根内任意目录直接跑即可,**不需要** `--project`。
|
|
49
|
+
|
|
50
|
+
**补传能做什么、不能做什么**:服务端对同一 change key 只保存**一个不可变包**。所以补传只适用于
|
|
51
|
+
「从未成功上传」与「字节完全一致的重试」两种情况。若该 change 已是 durable,`republish` 会读
|
|
52
|
+
`.harness/state/local/archive-packages/<key>.remote.json` 在**本地就判定**并返回:
|
|
53
|
+
字节一致 → `ARCHIVE_ALREADY_PUBLISHED`(exit 0,不重传);字节不同 →
|
|
54
|
+
`ARCHIVE_REMOTE_IMMUTABLE_CONFLICT`(exit 1,**不发起上传**)。
|
|
55
|
+
**已发布的归档无法从客户端补上知识条目**——注入 `candidates/knowledge.json` 必然改变字节,
|
|
56
|
+
服务端拒绝替换。要让旧归档产生知识条目,需要平台侧提供重新索引或归档版本化能力。
|
|
57
|
+
只想重试**同一个包**(例如知识索引失败)时加 `--no-knowledge-injection`,它按封存目录原样重建,字节与已存包一致。
|
|
44
58
|
|
|
45
59
|
## 关键规则
|
|
46
60
|
|
|
@@ -4,7 +4,7 @@ description: 6维度代码审查(架构/安全/规范/兼容/测试/性能)
|
|
|
4
4
|
.harness/context-index.json)和测试场景表,在隔离上下文运行。仅当用户显式调用 /harness-review 时使用;不得在
|
|
5
5
|
test 结束后自动接续执行。
|
|
6
6
|
---
|
|
7
|
-
<!-- generated by harness_deploy.py; core=
|
|
7
|
+
<!-- generated by harness_deploy.py; core=7fdcf0b62789ff65; overlay=none; agent=codex; do not edit -->
|
|
8
8
|
# harness-review — 代码审查
|
|
9
9
|
|
|
10
10
|
## Purpose
|
|
@@ -144,13 +144,23 @@ review 读取 `state-snapshot.json`(`harness_state.py` / state-layout-protocol
|
|
|
144
144
|
|
|
145
145
|
`events.ndjson` 为唯一事实源(schema_version 3,兼容读取 v1/v2;`note` 承载人类可读摘要);`logs/execution-log.md` 由 `harness_events.py` 渲染,**禁止用 Write/Edit 直接维护**。直接修改的内容会在 `phase.end` 或 finalize 时被完整重建覆盖,属于数据丢失;需要保留的详情必须进入事件 `note`。结构 → [[../protocols/report-pipeline-protocol.md|report-pipeline-protocol]]
|
|
146
146
|
|
|
147
|
+
**`phase.start` 由 `harness_gate.py begin` 写,不要再手工追加一次。** 两条同 `run-id` 的
|
|
148
|
+
`phase.start` 会让 `plan finalize` 以 `PHASE_START_DUPLICATE` 卡死,而且手工那次会先触发
|
|
149
|
+
auto-seal、把正在开始的 attempt 封成 `RECOVERED`。要补触发指令说明就带 `--note` 跑 `gate begin`。
|
|
150
|
+
(重复追加现已按 `(phase, run-id)` 判为幂等 no-op,但依赖它不如不写。)
|
|
151
|
+
|
|
147
152
|
```powershell
|
|
148
|
-
|
|
153
|
+
# 阶段开始:gate begin 负责,note 在这里给
|
|
154
|
+
python <skills-root>/scripts/harness_gate.py begin --change-dir ".harness/changes/<change-name>" --phase <phase> --note "<触发指令>"
|
|
155
|
+
# 阶段中的其他事件才用 append
|
|
156
|
+
python <skills-root>/scripts/harness_events.py append --change-dir ".harness/changes/<change-name>" --phase <phase> --type <command|issue|verification> --run-id <phase-run-id> --note "<摘要>"
|
|
149
157
|
```
|
|
150
158
|
|
|
151
159
|
> **脚本接线**:`harness_events.py append`;`harness_archive.py finalize`;`harness_preflight.py check`;`harness_ledger.py can-reuse`;`harness_service.py ensure/stop`(须 `--files`/`serviceStart.inputFiles`)。JSON 输出按 D13 护栏解读。
|
|
152
160
|
|
|
153
161
|
> **Task 4 §6.1 写入契约**:普通 `append` = 加锁 -> 追加一行 -> fsync -> 解锁,**不 load 历史、不渲染**(O(1),跨进程锁 `events.ndjson.lock`,UUID 用完整 `uuid4().hex` 无需去重扫描)。仅 `--type phase.end` append 在追加成功后渲染一次 `execution-log.md`;显式 `harness_events.py render` 随时从完整 events 重建;`harness_archive.py finalize` 在 collect 前强制 render 一次。高频 command append 期间 log 可能滞后,phase 边界保持最新。
|
|
154
162
|
|
|
155
|
-
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n
|
|
163
|
+
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n>`,不得覆盖或伪装成一次执行。**`attempt` 按 phase 全局递增,不是按 run-id**:一个 run-id 只绑定一个 attempt,重试必须「新 run-id + 下一个 attempt」两者同时换,只换其一会撞 `EVENT_ATTEMPT_CONFLICT` 或 `PHASE_ALREADY_CLOSED`。已发布的 plan 用 `harness_plan_finalize.py republish` 自动分配两者。
|
|
164
|
+
|
|
165
|
+
阶段跑得久(plan/run 常见)时用 `harness_context.py renew --project . --change <cn> --executor <tool>` 续租;租约到期本身不再阻断 `close`(同一 owner 的过期租约不构成冲突,收据里记 `leaseLapsed`),但续租能让 `view` 的状态如实反映在跑。跨工具继续执行时写 `--executor-tool <codex|claude-code|codebuddy|cursor>`,并在接棒事件写 `--handoff-from-tool` / `--handoff-reason`;也可由 `HUNTER_HARNESS_TOOL/AGENT/MODEL/RUN_ID` 环境变量统一注入。
|
|
156
166
|
> 片段:logging · phase=`review` · 事件:phase/decision/verification/issue/artifact
|
|
@@ -3,7 +3,7 @@ name: harness-run
|
|
|
3
3
|
description: 按变更簇执行 TDD 编码循环(RED→GREEN→REFACTOR→编译验证),逐变更簇实现计划中的任务。仅当用户显式调用
|
|
4
4
|
/harness-run 时使用;不得因用户提到编码/实现就自动触发,也不得被其他阶段 skill 自动接续。
|
|
5
5
|
---
|
|
6
|
-
<!-- generated by harness_deploy.py; core=
|
|
6
|
+
<!-- generated by harness_deploy.py; core=7fdcf0b62789ff65; overlay=none; agent=codex; do not edit -->
|
|
7
7
|
# harness-run — 需求编码
|
|
8
8
|
|
|
9
9
|
## Purpose
|
|
@@ -138,13 +138,23 @@ python <skills-root>/scripts/harness_test_guard.py record --project . --change-d
|
|
|
138
138
|
|
|
139
139
|
`events.ndjson` 为唯一事实源(schema_version 3,兼容读取 v1/v2;`note` 承载人类可读摘要);`logs/execution-log.md` 由 `harness_events.py` 渲染,**禁止用 Write/Edit 直接维护**。直接修改的内容会在 `phase.end` 或 finalize 时被完整重建覆盖,属于数据丢失;需要保留的详情必须进入事件 `note`。结构 → [[../protocols/report-pipeline-protocol.md|report-pipeline-protocol]]
|
|
140
140
|
|
|
141
|
+
**`phase.start` 由 `harness_gate.py begin` 写,不要再手工追加一次。** 两条同 `run-id` 的
|
|
142
|
+
`phase.start` 会让 `plan finalize` 以 `PHASE_START_DUPLICATE` 卡死,而且手工那次会先触发
|
|
143
|
+
auto-seal、把正在开始的 attempt 封成 `RECOVERED`。要补触发指令说明就带 `--note` 跑 `gate begin`。
|
|
144
|
+
(重复追加现已按 `(phase, run-id)` 判为幂等 no-op,但依赖它不如不写。)
|
|
145
|
+
|
|
141
146
|
```powershell
|
|
142
|
-
|
|
147
|
+
# 阶段开始:gate begin 负责,note 在这里给
|
|
148
|
+
python <skills-root>/scripts/harness_gate.py begin --change-dir ".harness/changes/<change-name>" --phase <phase> --note "<触发指令>"
|
|
149
|
+
# 阶段中的其他事件才用 append
|
|
150
|
+
python <skills-root>/scripts/harness_events.py append --change-dir ".harness/changes/<change-name>" --phase <phase> --type <command|issue|verification> --run-id <phase-run-id> --note "<摘要>"
|
|
143
151
|
```
|
|
144
152
|
|
|
145
153
|
> **脚本接线**:`harness_events.py append`;`harness_archive.py finalize`;`harness_preflight.py check`;`harness_ledger.py can-reuse`;`harness_service.py ensure/stop`(须 `--files`/`serviceStart.inputFiles`)。JSON 输出按 D13 护栏解读。
|
|
146
154
|
|
|
147
155
|
> **Task 4 §6.1 写入契约**:普通 `append` = 加锁 -> 追加一行 -> fsync -> 解锁,**不 load 历史、不渲染**(O(1),跨进程锁 `events.ndjson.lock`,UUID 用完整 `uuid4().hex` 无需去重扫描)。仅 `--type phase.end` append 在追加成功后渲染一次 `execution-log.md`;显式 `harness_events.py render` 随时从完整 events 重建;`harness_archive.py finalize` 在 collect 前强制 render 一次。高频 command append 期间 log 可能滞后,phase 边界保持最新。
|
|
148
156
|
|
|
149
|
-
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n
|
|
157
|
+
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n>`,不得覆盖或伪装成一次执行。**`attempt` 按 phase 全局递增,不是按 run-id**:一个 run-id 只绑定一个 attempt,重试必须「新 run-id + 下一个 attempt」两者同时换,只换其一会撞 `EVENT_ATTEMPT_CONFLICT` 或 `PHASE_ALREADY_CLOSED`。已发布的 plan 用 `harness_plan_finalize.py republish` 自动分配两者。
|
|
158
|
+
|
|
159
|
+
阶段跑得久(plan/run 常见)时用 `harness_context.py renew --project . --change <cn> --executor <tool>` 续租;租约到期本身不再阻断 `close`(同一 owner 的过期租约不构成冲突,收据里记 `leaseLapsed`),但续租能让 `view` 的状态如实反映在跑。跨工具继续执行时写 `--executor-tool <codex|claude-code|codebuddy|cursor>`,并在接棒事件写 `--handoff-from-tool` / `--handoff-reason`;也可由 `HUNTER_HARNESS_TOOL/AGENT/MODEL/RUN_ID` 环境变量统一注入。
|
|
150
160
|
> 片段:logging · phase=`run`
|
|
@@ -3,7 +3,7 @@ name: harness-submit
|
|
|
3
3
|
description: 最终提交封装:验证→中文 commit→提交/推送;worktree 模式含 --no-ff 合并回主分支。仅当用户显式调用
|
|
4
4
|
/harness-submit(或 /harness-merge 重入合并段)时使用;用户口头说'提交/commit/push'时必须先确认,不得自动触发。
|
|
5
5
|
---
|
|
6
|
-
<!-- generated by harness_deploy.py; core=
|
|
6
|
+
<!-- generated by harness_deploy.py; core=7fdcf0b62789ff65; overlay=none; agent=codex; do not edit -->
|
|
7
7
|
# harness-submit — 最终提交(含 worktree 合并)
|
|
8
8
|
|
|
9
9
|
## Purpose
|
|
@@ -160,13 +160,23 @@ git 经 PowerShell;commit/报告不得含明文密钥。遵循 `../protocols/s
|
|
|
160
160
|
|
|
161
161
|
`events.ndjson` 为唯一事实源(schema_version 3,兼容读取 v1/v2;`note` 承载人类可读摘要);`logs/execution-log.md` 由 `harness_events.py` 渲染,**禁止用 Write/Edit 直接维护**。直接修改的内容会在 `phase.end` 或 finalize 时被完整重建覆盖,属于数据丢失;需要保留的详情必须进入事件 `note`。结构 → [[../protocols/report-pipeline-protocol.md|report-pipeline-protocol]]
|
|
162
162
|
|
|
163
|
+
**`phase.start` 由 `harness_gate.py begin` 写,不要再手工追加一次。** 两条同 `run-id` 的
|
|
164
|
+
`phase.start` 会让 `plan finalize` 以 `PHASE_START_DUPLICATE` 卡死,而且手工那次会先触发
|
|
165
|
+
auto-seal、把正在开始的 attempt 封成 `RECOVERED`。要补触发指令说明就带 `--note` 跑 `gate begin`。
|
|
166
|
+
(重复追加现已按 `(phase, run-id)` 判为幂等 no-op,但依赖它不如不写。)
|
|
167
|
+
|
|
163
168
|
```powershell
|
|
164
|
-
|
|
169
|
+
# 阶段开始:gate begin 负责,note 在这里给
|
|
170
|
+
python <skills-root>/scripts/harness_gate.py begin --change-dir ".harness/changes/<change-name>" --phase <phase> --note "<触发指令>"
|
|
171
|
+
# 阶段中的其他事件才用 append
|
|
172
|
+
python <skills-root>/scripts/harness_events.py append --change-dir ".harness/changes/<change-name>" --phase <phase> --type <command|issue|verification> --run-id <phase-run-id> --note "<摘要>"
|
|
165
173
|
```
|
|
166
174
|
|
|
167
175
|
> **脚本接线**:`harness_events.py append`;`harness_archive.py finalize`;`harness_preflight.py check`;`harness_ledger.py can-reuse`;`harness_service.py ensure/stop`(须 `--files`/`serviceStart.inputFiles`)。JSON 输出按 D13 护栏解读。
|
|
168
176
|
|
|
169
177
|
> **Task 4 §6.1 写入契约**:普通 `append` = 加锁 -> 追加一行 -> fsync -> 解锁,**不 load 历史、不渲染**(O(1),跨进程锁 `events.ndjson.lock`,UUID 用完整 `uuid4().hex` 无需去重扫描)。仅 `--type phase.end` append 在追加成功后渲染一次 `execution-log.md`;显式 `harness_events.py render` 随时从完整 events 重建;`harness_archive.py finalize` 在 collect 前强制 render 一次。高频 command append 期间 log 可能滞后,phase 边界保持最新。
|
|
170
178
|
|
|
171
|
-
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n
|
|
179
|
+
每个阶段的 `phase.start` 与对应 `phase.end` 必须复用同一 `--run-id` / `--attempt`;阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`。重试同一阶段时生成新的 run-id 并增加 `--attempt <n>`,不得覆盖或伪装成一次执行。**`attempt` 按 phase 全局递增,不是按 run-id**:一个 run-id 只绑定一个 attempt,重试必须「新 run-id + 下一个 attempt」两者同时换,只换其一会撞 `EVENT_ATTEMPT_CONFLICT` 或 `PHASE_ALREADY_CLOSED`。已发布的 plan 用 `harness_plan_finalize.py republish` 自动分配两者。
|
|
180
|
+
|
|
181
|
+
阶段跑得久(plan/run 常见)时用 `harness_context.py renew --project . --change <cn> --executor <tool>` 续租;租约到期本身不再阻断 `close`(同一 owner 的过期租约不构成冲突,收据里记 `leaseLapsed`),但续租能让 `view` 的状态如实反映在跑。跨工具继续执行时写 `--executor-tool <codex|claude-code|codebuddy|cursor>`,并在接棒事件写 `--handoff-from-tool` / `--handoff-reason`;也可由 `HUNTER_HARNESS_TOOL/AGENT/MODEL/RUN_ID` 环境变量统一注入。
|
|
172
182
|
> 片段:logging · phase=`submit`/`merge`
|
|
@@ -4,7 +4,7 @@ description: Use when the user asks to synchronize, refresh, or validate Harness
|
|
|
4
4
|
metadata, adapters, remote knowledge ownership, instruction entrypoints,
|
|
5
5
|
config origins, or CodeGraph status.
|
|
6
6
|
---
|
|
7
|
-
<!-- generated by harness_deploy.py; core=
|
|
7
|
+
<!-- generated by harness_deploy.py; core=7fdcf0b62789ff65; overlay=none; agent=codex; do not edit -->
|
|
8
8
|
# harness-sync
|
|
9
9
|
|
|
10
10
|
## Purpose
|