okstra 0.231.1 → 0.232.0
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/README.md +12 -0
- package/dist/cli-registry.mjs +6 -0
- package/dist/cli-registry.mjs.map +1 -1
- package/dist/commands/memory/memory.mjs +9 -55
- package/dist/commands/memory/memory.mjs.map +1 -1
- package/dist/lib/memory-auto-register.d.mts +1 -0
- package/dist/lib/memory-auto-register.mjs +113 -0
- package/dist/lib/memory-auto-register.mjs.map +1 -0
- package/dist/lib/memory-entry-contract.d.mts +12 -0
- package/dist/lib/memory-entry-contract.mjs +46 -0
- package/dist/lib/memory-entry-contract.mjs.map +1 -0
- package/dist/lib/python-helper.d.mts +1 -1
- package/dist/lib/python-helper.mjs +3 -2
- package/dist/lib/python-helper.mjs.map +1 -1
- package/dist/lib/record-secrets.d.mts +1 -0
- package/dist/lib/record-secrets.mjs +28 -0
- package/dist/lib/record-secrets.mjs.map +1 -0
- package/dist/lib/types.d.mts +2 -1
- package/docs/architecture/storage-model.md +127 -0
- package/docs/architecture.md +7 -0
- package/docs/cli.md +234 -0
- package/docs/project-structure-overview.md +2 -1
- package/package.json +1 -1
- package/runtime/BUILD.json +2 -2
- package/runtime/agents/common.json +8 -8
- package/runtime/agents/roles/verifier.json +2 -4
- package/runtime/bin/okstra-provider-exec.py +10 -4
- package/runtime/prompts/coding-preflight/architectures/hexagonal.md +23 -30
- package/runtime/prompts/coding-preflight/authoring-checks.md +17 -0
- package/runtime/prompts/coding-preflight/clean-code.md +85 -281
- package/runtime/prompts/coding-preflight/frameworks/node-server.md +2 -3
- package/runtime/prompts/coding-preflight/languages/java.md +3 -5
- package/runtime/prompts/coding-preflight/languages/javascript-typescript.md +5 -12
- package/runtime/prompts/coding-preflight/languages/kotlin.md +2 -4
- package/runtime/prompts/coding-preflight/languages/python.md +2 -7
- package/runtime/prompts/coding-preflight/languages/rust.md +2 -8
- package/runtime/prompts/coding-preflight/languages/sql.md +2 -4
- package/runtime/prompts/coding-preflight/overview.md +25 -38
- package/runtime/prompts/coding-preflight/review-rules.json +30 -0
- package/runtime/prompts/duties/code-reviewer.json +15 -18
- package/runtime/prompts/duties/implementation-verifier.json +1 -0
- package/runtime/prompts/profiles/_coding-conventions-preflight.md +41 -6
- package/runtime/python/okstra_ctl/agent/invocation.py +121 -36
- package/runtime/python/okstra_ctl/agent/prompt_cli/cli.py +6 -0
- package/runtime/python/okstra_ctl/agent/prompt_cli/materialize.py +30 -4
- package/runtime/python/okstra_ctl/agent/standalone.py +3 -0
- package/runtime/python/okstra_ctl/analysis_packet.py +2 -19
- package/runtime/python/okstra_ctl/attempt_evidence.py +10 -0
- package/runtime/python/okstra_ctl/code_review_census.py +71 -7
- package/runtime/python/okstra_ctl/code_review_coverage.py +4 -274
- package/runtime/python/okstra_ctl/code_review_integrity.py +154 -0
- package/runtime/python/okstra_ctl/code_review_invocation.py +212 -0
- package/runtime/python/okstra_ctl/code_review_scope.py +36 -9
- package/runtime/python/okstra_ctl/code_review_target.py +85 -11
- package/runtime/python/okstra_ctl/consumers.py +12 -9
- package/runtime/python/okstra_ctl/design_review.py +207 -0
- package/runtime/python/okstra_ctl/design_review_evidence.py +297 -0
- package/runtime/python/okstra_ctl/doctor.py +8 -25
- package/runtime/python/okstra_ctl/doyouknow/__init__.py +9 -0
- package/runtime/python/okstra_ctl/doyouknow/cli.py +91 -0
- package/runtime/python/okstra_ctl/doyouknow/config.py +58 -0
- package/runtime/python/okstra_ctl/doyouknow/contracts.py +184 -0
- package/runtime/python/okstra_ctl/doyouknow/identity.py +22 -0
- package/runtime/python/okstra_ctl/doyouknow/paths.py +22 -0
- package/runtime/python/okstra_ctl/doyouknow/pending.py +72 -0
- package/runtime/python/okstra_ctl/doyouknow/registration.py +260 -0
- package/runtime/python/okstra_ctl/doyouknow/source_contracts.py +128 -0
- package/runtime/python/okstra_ctl/doyouknow/source_store.py +172 -0
- package/runtime/python/okstra_ctl/doyouknow/sources/io.py +36 -0
- package/runtime/python/okstra_ctl/doyouknow/sources/manager.py +261 -0
- package/runtime/python/okstra_ctl/doyouknow/sources/memory.py +179 -0
- package/runtime/python/okstra_ctl/doyouknow/sources/task.py +95 -0
- package/runtime/python/okstra_ctl/doyouknow/store.py +260 -0
- package/runtime/python/okstra_ctl/doyouknow/sync.py +58 -0
- package/runtime/python/okstra_ctl/execution_mutation_audit.py +18 -2
- package/runtime/python/okstra_ctl/file_size_cap.py +3 -4
- package/runtime/python/okstra_ctl/json_boundary.py +21 -2
- package/runtime/python/okstra_ctl/jsonl.py +21 -0
- package/runtime/python/okstra_ctl/manager_cli.py +26 -0
- package/runtime/python/okstra_ctl/manager_registration.py +29 -0
- package/runtime/python/okstra_ctl/manager_sync.py +3 -22
- package/runtime/python/okstra_ctl/model_io/renderers.py +13 -2
- package/runtime/python/okstra_ctl/model_io_cli.py +7 -1
- package/runtime/python/okstra_ctl/phases/final_verification/validation.py +20 -0
- package/runtime/python/okstra_ctl/phases/implementation/instructions/_implementation-deliverable.md +10 -0
- package/runtime/python/okstra_ctl/phases/implementation/instructions/_implementation-diff-review.md +19 -6
- package/runtime/python/okstra_ctl/phases/implementation/instructions/_implementation-self-check.md +5 -3
- package/runtime/python/okstra_ctl/phases/implementation/instructions/_implementation-verifier.md +38 -10
- package/runtime/python/okstra_ctl/phases/implementation/report_assets/implementation.template.html +13 -0
- package/runtime/python/okstra_ctl/phases/implementation/spec.md +25 -0
- package/runtime/python/okstra_ctl/phases/implementation/validation.py +6 -0
- package/runtime/python/okstra_ctl/phases/implementation_planning/authoring.py +4 -0
- package/runtime/python/okstra_ctl/phases/implementation_planning/profile.md +1 -1
- package/runtime/python/okstra_ctl/report_assembly.py +15 -0
- package/runtime/python/okstra_ctl/report_narrative.py +3 -0
- package/runtime/python/okstra_ctl/report_synthesis_packet.py +11 -2
- package/runtime/python/okstra_ctl/review_policy.py +359 -0
- package/runtime/python/okstra_ctl/review_verdicts.py +291 -0
- package/runtime/python/okstra_ctl/run.py +24 -1
- package/runtime/python/okstra_ctl/stage_done.py +11 -0
- package/runtime/python/okstra_ctl/standalone_review_audit.py +28 -0
- package/runtime/python/okstra_ctl/verification_generation.py +23 -1
- package/runtime/python/okstra_ctl/verification_input_types.py +15 -1
- package/runtime/python/okstra_ctl/verification_inputs.py +28 -6
- package/runtime/python/okstra_ctl/verification_invocation.py +7 -1
- package/runtime/python/okstra_ctl/verification_target.py +2 -0
- package/runtime/python/okstra_ctl/worker_prompt_headers.py +24 -9
- package/runtime/python/okstra_ctl/worker_runner.py +66 -10
- package/runtime/python/okstra_ctl/write_policy.py +34 -1
- package/runtime/python/okstra_project/__init__.py +2 -0
- package/runtime/python/okstra_project/dirs.py +18 -0
- package/runtime/python/okstra_project/state.py +25 -20
- package/runtime/schemas/doyouknow-config-v1.schema.json +20 -0
- package/runtime/schemas/doyouknow-v1.schema.json +65 -0
- package/runtime/schemas/final-report-v2.0.schema.json +157 -0
- package/runtime/schemas/final-report-v3.0.schema.json +157 -0
- package/runtime/schemas/memory-entry-contract-v1.json +11 -0
- package/runtime/schemas/record-secret-patterns.json +41 -0
- package/runtime/skills/okstra-code-review/SKILL.md +71 -123
- package/runtime/skills/okstra-code-review/references/census-rules.md +29 -105
- package/runtime/skills/okstra-code-review/references/review-calibration.md +34 -54
- package/runtime/skills/okstra-manager/SKILL.md +7 -0
- package/runtime/skills/okstra-memory/SKILL.md +7 -0
- package/runtime/templates/implementation-worker-preamble.md +4 -9
- package/runtime/templates/worker-prompt-preamble.md +4 -12
- package/runtime/validators/validate-run.py +14 -2
|
@@ -12,7 +12,7 @@ from pathlib import Path
|
|
|
12
12
|
from pathlib import PurePosixPath
|
|
13
13
|
import re
|
|
14
14
|
import tempfile
|
|
15
|
-
from typing import Any, Iterator, Literal, Mapping, get_args
|
|
15
|
+
from typing import TYPE_CHECKING, Any, Iterator, Literal, Mapping, get_args
|
|
16
16
|
|
|
17
17
|
from ..domain.role import RoleCatalogError, role_for_duty
|
|
18
18
|
from ..verification_invocation import (
|
|
@@ -26,6 +26,9 @@ from ..json_boundary import (
|
|
|
26
26
|
load_owned_object,
|
|
27
27
|
)
|
|
28
28
|
|
|
29
|
+
if TYPE_CHECKING:
|
|
30
|
+
from ..code_review_invocation import CodeReviewInputs
|
|
31
|
+
|
|
29
32
|
|
|
30
33
|
AgentAudience = Literal[
|
|
31
34
|
"lead",
|
|
@@ -190,6 +193,7 @@ class AgentInvocationRequest:
|
|
|
190
193
|
# dispatch has used it. See `_publish_or_reuse`.
|
|
191
194
|
replace_undispatched: bool = False
|
|
192
195
|
verification_generation: VerificationGenerationReference | None = None
|
|
196
|
+
review_inputs: CodeReviewInputs | None = None
|
|
193
197
|
|
|
194
198
|
|
|
195
199
|
@dataclass(frozen=True)
|
|
@@ -648,6 +652,9 @@ def prepare_agent_invocation(
|
|
|
648
652
|
"""Materialize one immutable prompt and its adjacent completion metadata."""
|
|
649
653
|
request = _bind_generation_request(request)
|
|
650
654
|
_validate_request(request)
|
|
655
|
+
from ..code_review_invocation import bind_review_invocation
|
|
656
|
+
|
|
657
|
+
request = bind_review_invocation(request)
|
|
651
658
|
materialized = _materialize(request)
|
|
652
659
|
if materialized.reservation_root is None:
|
|
653
660
|
with _exclusive_lock(_prompt_lock_path(request.prompt_path)):
|
|
@@ -944,8 +951,9 @@ def publish_standalone_completion(
|
|
|
944
951
|
invocation_id=invocation_id,
|
|
945
952
|
metadata_path=canonical_metadata,
|
|
946
953
|
)
|
|
954
|
+
review_audit = _complete_standalone_review_audit(project_root, canonical_metadata)
|
|
947
955
|
completion = {
|
|
948
|
-
"schemaVersion": 1,
|
|
956
|
+
"schemaVersion": 2 if review_audit is not None else 1,
|
|
949
957
|
"purpose": purpose,
|
|
950
958
|
"invocationId": invocation_id,
|
|
951
959
|
"status": "completed",
|
|
@@ -962,10 +970,34 @@ def publish_standalone_completion(
|
|
|
962
970
|
"resultEnvelopeDigest": _sha256(envelope_bytes),
|
|
963
971
|
"completedAt": completed,
|
|
964
972
|
}
|
|
973
|
+
if review_audit is not None:
|
|
974
|
+
completion["reviewAuditDigest"] = review_audit
|
|
965
975
|
_publish_exclusive(completion_path, _canonical_json(completion))
|
|
966
976
|
return completion_path
|
|
967
977
|
|
|
968
978
|
|
|
979
|
+
def _complete_standalone_review_audit(project_root: Path, metadata_path: Path) -> str | None:
|
|
980
|
+
from ..code_review_invocation import complete_review_audit
|
|
981
|
+
|
|
982
|
+
try:
|
|
983
|
+
return complete_review_audit(
|
|
984
|
+
project_root, metadata_path,
|
|
985
|
+
_load_json_object(metadata_path, "agent invocation metadata"),
|
|
986
|
+
)
|
|
987
|
+
except (OSError, ValueError, KeyError) as exc:
|
|
988
|
+
raise AgentInvocationError(f"standalone review integrity failed: {exc}") from exc
|
|
989
|
+
|
|
990
|
+
|
|
991
|
+
@dataclass(frozen=True)
|
|
992
|
+
class _StandaloneCompletion:
|
|
993
|
+
payload: Mapping[str, Any]
|
|
994
|
+
invocation_id: str
|
|
995
|
+
root: Path
|
|
996
|
+
metadata_path: Path
|
|
997
|
+
result_path: Path
|
|
998
|
+
completion_path: Path
|
|
999
|
+
|
|
1000
|
+
|
|
969
1001
|
def verify_standalone_completion(
|
|
970
1002
|
completion_path: Path,
|
|
971
1003
|
*,
|
|
@@ -973,6 +1005,49 @@ def verify_standalone_completion(
|
|
|
973
1005
|
expected_purpose: str,
|
|
974
1006
|
) -> VerifiedStandaloneResult:
|
|
975
1007
|
"""Verify a completion marker and return its already-read immutable body."""
|
|
1008
|
+
context = _read_standalone_completion(completion_path, project_root, expected_purpose)
|
|
1009
|
+
with _exclusive_lock(context.root / f"{context.invocation_id}.result.publish.lock"):
|
|
1010
|
+
authority_id, _authority_root, canonical_metadata = _standalone_authority(
|
|
1011
|
+
project_root=project_root,
|
|
1012
|
+
purpose=expected_purpose,
|
|
1013
|
+
metadata_path=context.metadata_path,
|
|
1014
|
+
)
|
|
1015
|
+
if authority_id != context.invocation_id:
|
|
1016
|
+
raise AgentInvocationError("standalone completion identity does not match")
|
|
1017
|
+
envelope_bytes, returned_body = _verified_result_envelope(
|
|
1018
|
+
context.result_path,
|
|
1019
|
+
project_root=project_root,
|
|
1020
|
+
purpose=expected_purpose,
|
|
1021
|
+
invocation_id=context.invocation_id,
|
|
1022
|
+
metadata_path=canonical_metadata,
|
|
1023
|
+
)
|
|
1024
|
+
digest = _sha256(envelope_bytes)
|
|
1025
|
+
if context.payload.get("resultEnvelopeDigest") != digest:
|
|
1026
|
+
raise AgentInvocationError("result envelope digest does not match completion")
|
|
1027
|
+
from ..code_review_invocation import verify_review_completion
|
|
1028
|
+
|
|
1029
|
+
try:
|
|
1030
|
+
verify_review_completion(
|
|
1031
|
+
project_root, canonical_metadata,
|
|
1032
|
+
_load_json_object(canonical_metadata, "agent invocation metadata"),
|
|
1033
|
+
context.payload,
|
|
1034
|
+
)
|
|
1035
|
+
except (OSError, ValueError, KeyError) as exc:
|
|
1036
|
+
raise AgentInvocationError(f"standalone review integrity is invalid: {exc}") from exc
|
|
1037
|
+
return VerifiedStandaloneResult(
|
|
1038
|
+
purpose=expected_purpose,
|
|
1039
|
+
invocation_id=context.invocation_id,
|
|
1040
|
+
prompt_metadata_path=context.metadata_path,
|
|
1041
|
+
result_path=context.result_path,
|
|
1042
|
+
completion_path=context.completion_path,
|
|
1043
|
+
result_envelope_digest=digest,
|
|
1044
|
+
returned_body=returned_body,
|
|
1045
|
+
)
|
|
1046
|
+
|
|
1047
|
+
|
|
1048
|
+
def _read_standalone_completion(
|
|
1049
|
+
completion_path: Path, project_root: Path, expected_purpose: str,
|
|
1050
|
+
) -> _StandaloneCompletion:
|
|
976
1051
|
_validate_slug(expected_purpose, "standalone purpose")
|
|
977
1052
|
completion = _load_json_object(completion_path, "standalone completion")
|
|
978
1053
|
expected_keys = {
|
|
@@ -985,7 +1060,9 @@ def verify_standalone_completion(
|
|
|
985
1060
|
"resultEnvelopeDigest",
|
|
986
1061
|
"completedAt",
|
|
987
1062
|
}
|
|
988
|
-
if
|
|
1063
|
+
if completion.get("schemaVersion") == 2:
|
|
1064
|
+
expected_keys.add("reviewAuditDigest")
|
|
1065
|
+
if set(completion) != expected_keys or completion.get("schemaVersion") not in (1, 2):
|
|
989
1066
|
raise AgentInvocationError("standalone completion schema is invalid")
|
|
990
1067
|
if completion.get("purpose") != expected_purpose:
|
|
991
1068
|
raise AgentInvocationError("standalone completion purpose does not match")
|
|
@@ -1013,32 +1090,8 @@ def verify_standalone_completion(
|
|
|
1013
1090
|
must_exist=True,
|
|
1014
1091
|
):
|
|
1015
1092
|
raise AgentInvocationError("standalone completion result path does not match")
|
|
1016
|
-
|
|
1017
|
-
|
|
1018
|
-
project_root=project_root,
|
|
1019
|
-
purpose=expected_purpose,
|
|
1020
|
-
metadata_path=metadata_path,
|
|
1021
|
-
)
|
|
1022
|
-
if authority_id != invocation_id:
|
|
1023
|
-
raise AgentInvocationError("standalone completion identity does not match")
|
|
1024
|
-
envelope_bytes, returned_body = _verified_result_envelope(
|
|
1025
|
-
result_path,
|
|
1026
|
-
project_root=project_root,
|
|
1027
|
-
purpose=expected_purpose,
|
|
1028
|
-
invocation_id=invocation_id,
|
|
1029
|
-
metadata_path=canonical_metadata,
|
|
1030
|
-
)
|
|
1031
|
-
digest = _sha256(envelope_bytes)
|
|
1032
|
-
if completion.get("resultEnvelopeDigest") != digest:
|
|
1033
|
-
raise AgentInvocationError("result envelope digest does not match completion")
|
|
1034
|
-
return VerifiedStandaloneResult(
|
|
1035
|
-
purpose=expected_purpose,
|
|
1036
|
-
invocation_id=invocation_id,
|
|
1037
|
-
prompt_metadata_path=metadata_path,
|
|
1038
|
-
result_path=result_path,
|
|
1039
|
-
completion_path=canonical_completion,
|
|
1040
|
-
result_envelope_digest=digest,
|
|
1041
|
-
returned_body=returned_body,
|
|
1093
|
+
return _StandaloneCompletion(
|
|
1094
|
+
completion, invocation_id, root, metadata_path, result_path, canonical_completion,
|
|
1042
1095
|
)
|
|
1043
1096
|
|
|
1044
1097
|
|
|
@@ -1090,6 +1143,8 @@ def _verified_result_envelope(
|
|
|
1090
1143
|
invocation_id: str,
|
|
1091
1144
|
metadata_path: Path,
|
|
1092
1145
|
) -> tuple[bytes, str]:
|
|
1146
|
+
project_root = project_root.resolve(strict=True)
|
|
1147
|
+
result_path = result_path.parent.resolve(strict=True) / result_path.name
|
|
1093
1148
|
try:
|
|
1094
1149
|
body = result_path.read_bytes()
|
|
1095
1150
|
envelope = load_external_json(
|
|
@@ -1184,11 +1239,30 @@ def _bind_generation_request(request: AgentInvocationRequest) -> AgentInvocation
|
|
|
1184
1239
|
try:
|
|
1185
1240
|
manifest = _load_request_manifest(request)
|
|
1186
1241
|
reference = _typed_generation_reference(request)
|
|
1187
|
-
|
|
1242
|
+
generation = read_invocation_generation(
|
|
1188
1243
|
request.project_root, request.run_manifest_path, manifest,
|
|
1189
1244
|
reference, request.audience,
|
|
1190
1245
|
)
|
|
1191
|
-
|
|
1246
|
+
bound = _with_generation_anchors(request, reference)
|
|
1247
|
+
if generation is not None and "designReview" in generation:
|
|
1248
|
+
from ..design_review import render_design_review
|
|
1249
|
+
|
|
1250
|
+
supplement = render_design_review(generation["designReview"])
|
|
1251
|
+
from ..design_review_evidence import prior_design_review_prompt
|
|
1252
|
+
|
|
1253
|
+
supplement += prior_design_review_prompt(
|
|
1254
|
+
request.project_root, request.run_manifest_path,
|
|
1255
|
+
request.role_execution_ref, generation["generation"],
|
|
1256
|
+
request.invocation_ref, request.attempt,
|
|
1257
|
+
)
|
|
1258
|
+
body = bound.instruction.body
|
|
1259
|
+
if not body.rstrip().endswith(supplement.rstrip()):
|
|
1260
|
+
if "\n## Frozen Design Review\n" in body:
|
|
1261
|
+
raise ValueError("design review body differs from sealed generation")
|
|
1262
|
+
bound = replace(bound, instruction=replace(
|
|
1263
|
+
bound.instruction, body=body.rstrip() + "\n" + supplement,
|
|
1264
|
+
))
|
|
1265
|
+
return bound
|
|
1192
1266
|
except (OSError, ValueError) as exc:
|
|
1193
1267
|
raise AgentInvocationError(str(exc)) from exc
|
|
1194
1268
|
|
|
@@ -1973,6 +2047,9 @@ def _verify_prompt_file(
|
|
|
1973
2047
|
errors.extend(_verify_model_header(metadata, body))
|
|
1974
2048
|
errors.extend(_verify_execution_delivery(metadata, path, body))
|
|
1975
2049
|
errors.extend(_verify_generation_reference(metadata, project_root, body))
|
|
2050
|
+
from ..code_review_invocation import review_policy_errors_for_prompt
|
|
2051
|
+
|
|
2052
|
+
errors.extend(review_policy_errors_for_prompt(project_root, metadata, body))
|
|
1976
2053
|
return errors
|
|
1977
2054
|
|
|
1978
2055
|
|
|
@@ -2010,6 +2087,16 @@ def reserved_generation_reference(
|
|
|
2010
2087
|
The reservation names the metadata, so a reader does not take a generation
|
|
2011
2088
|
file on the strength of the path its dispatch kind computes.
|
|
2012
2089
|
"""
|
|
2090
|
+
metadata_path = reserved_invocation_metadata_path(project_root, manifest, invocation_ref, attempt)
|
|
2091
|
+
if metadata_path is None:
|
|
2092
|
+
return None
|
|
2093
|
+
metadata = _load_json_object(metadata_path, "agent invocation metadata")
|
|
2094
|
+
return metadata_generation_reference(metadata)
|
|
2095
|
+
|
|
2096
|
+
|
|
2097
|
+
def reserved_invocation_metadata_path(
|
|
2098
|
+
project_root: Path, manifest: Mapping[str, object], invocation_ref: str, attempt: int,
|
|
2099
|
+
) -> Path | None:
|
|
2013
2100
|
contract = manifest.get("agentContract")
|
|
2014
2101
|
if not isinstance(contract, Mapping):
|
|
2015
2102
|
return None
|
|
@@ -2020,17 +2107,15 @@ def reserved_generation_reference(
|
|
|
2020
2107
|
if not path.is_file():
|
|
2021
2108
|
return None
|
|
2022
2109
|
reservation = _load_json_object(path, "invocation reservation")
|
|
2023
|
-
|
|
2024
|
-
|
|
2025
|
-
"agent invocation metadata",
|
|
2026
|
-
)
|
|
2110
|
+
metadata_path = _project_path(project_root, reservation.get("metadataPath"), must_exist=True)
|
|
2111
|
+
metadata = _load_json_object(metadata_path, "agent invocation metadata")
|
|
2027
2112
|
for payload in (reservation, metadata):
|
|
2028
2113
|
if (payload.get("invocationRef"), payload.get("attempt")) != (
|
|
2029
2114
|
invocation_ref, attempt):
|
|
2030
2115
|
raise AgentInvocationError(
|
|
2031
2116
|
f"invocation reservation {path} names another invocation attempt"
|
|
2032
2117
|
)
|
|
2033
|
-
return
|
|
2118
|
+
return metadata_path
|
|
2034
2119
|
|
|
2035
2120
|
|
|
2036
2121
|
def _attempt_invocation_id(invocation_ref: str, attempt: int) -> str:
|
|
@@ -129,6 +129,12 @@ def _add_materialize_parser(commands: argparse._SubParsersAction) -> None:
|
|
|
129
129
|
materialize.add_argument("--model-role", help=standalone_only)
|
|
130
130
|
materialize.add_argument("--model", default="", help=standalone_only)
|
|
131
131
|
materialize.add_argument("--purpose", help=standalone_only)
|
|
132
|
+
materialize.add_argument("--review-policy", help=standalone_only + "; frozen shared code-review policy")
|
|
133
|
+
materialize.add_argument("--review-census", help=standalone_only + "; assigned code-review census")
|
|
134
|
+
materialize.add_argument("--review-base", help=standalone_only + "; fixed review base")
|
|
135
|
+
materialize.add_argument("--review-head", help=standalone_only + "; fixed review head")
|
|
136
|
+
materialize.add_argument("--review-resource", action="append", default=[],
|
|
137
|
+
help=standalone_only + "; additional frozen pack-relative resource")
|
|
132
138
|
materialize.add_argument(
|
|
133
139
|
"--replace-undispatched",
|
|
134
140
|
action="store_true",
|
|
@@ -32,6 +32,7 @@ from ..invocation import (
|
|
|
32
32
|
)
|
|
33
33
|
from ...assignment_environment import load_assignment_context
|
|
34
34
|
from ..standalone import StandaloneInvocationRequest, prepare_standalone_invocation
|
|
35
|
+
from ...code_review_invocation import CodeReviewInputs
|
|
35
36
|
from ...path_hints import hydrate_active_run_context
|
|
36
37
|
from ...worker_prompt_headers import worker_prompt_headers
|
|
37
38
|
from ...worker_prompt_contract import (
|
|
@@ -134,6 +135,9 @@ def _materialize_run(
|
|
|
134
135
|
"model-role": args.model_role,
|
|
135
136
|
"model": args.model,
|
|
136
137
|
"purpose": args.purpose,
|
|
138
|
+
**{f"review-{name}": getattr(args, f"review_{name}", None)
|
|
139
|
+
for name in ("policy", "census", "base", "head")},
|
|
140
|
+
"review-resource": getattr(args, "review_resource", ()),
|
|
137
141
|
}
|
|
138
142
|
present = [name for name, value in forbidden.items() if value]
|
|
139
143
|
if present:
|
|
@@ -922,10 +926,9 @@ def _roster_result_path(
|
|
|
922
926
|
return None
|
|
923
927
|
|
|
924
928
|
|
|
925
|
-
def
|
|
926
|
-
args: argparse.Namespace,
|
|
927
|
-
|
|
928
|
-
) -> PreparedAgentInvocation:
|
|
929
|
+
def _standalone_input_paths(
|
|
930
|
+
args: argparse.Namespace, project_root: Path,
|
|
931
|
+
) -> tuple[Path, Path]:
|
|
929
932
|
required = {
|
|
930
933
|
"host-runtime": args.host_runtime,
|
|
931
934
|
"provider": args.provider,
|
|
@@ -967,15 +970,38 @@ def _materialize_standalone(
|
|
|
967
970
|
expected_prompt = root / f"{args.invocation_id}.prompt.md"
|
|
968
971
|
instruction_path = _exact_path(args.instruction, expected_instruction, True, "instruction")
|
|
969
972
|
prompt_path = _exact_path(args.prompt, expected_prompt, False, "prompt")
|
|
973
|
+
return instruction_path, prompt_path
|
|
974
|
+
|
|
975
|
+
|
|
976
|
+
def _materialize_standalone(
|
|
977
|
+
args: argparse.Namespace,
|
|
978
|
+
project_root: Path,
|
|
979
|
+
) -> PreparedAgentInvocation:
|
|
980
|
+
instruction_path, prompt_path = _standalone_input_paths(args, project_root)
|
|
970
981
|
assignment_context = load_assignment_context(
|
|
971
982
|
host_runtime=args.host_runtime,
|
|
972
983
|
terminal_backend=args.terminal_backend or detect_terminal_backend(),
|
|
973
984
|
execution_provider_ids=(args.provider,),
|
|
974
985
|
include_native_provider=False,
|
|
975
986
|
)
|
|
987
|
+
review_values = [getattr(args, f"review_{name}", None)
|
|
988
|
+
for name in ("policy", "census", "base", "head")]
|
|
989
|
+
review_inputs = None
|
|
990
|
+
if args.audience == "code-reviewer" or any(review_values):
|
|
991
|
+
if args.audience != "code-reviewer" or not all(review_values):
|
|
992
|
+
raise AgentPromptCliError(
|
|
993
|
+
"code-reviewer requires --review-policy, --review-census, --review-base and --review-head"
|
|
994
|
+
)
|
|
995
|
+
review_inputs = CodeReviewInputs(
|
|
996
|
+
_project_input(project_root, args.review_policy, "review policy"),
|
|
997
|
+
_project_input(project_root, args.review_census, "review census"),
|
|
998
|
+
args.review_base, args.review_head,
|
|
999
|
+
tuple(getattr(args, "review_resource", ())),
|
|
1000
|
+
)
|
|
976
1001
|
return prepare_standalone_invocation(StandaloneInvocationRequest(
|
|
977
1002
|
project_root=project_root, invocation_id=args.invocation_id, audience=args.audience,
|
|
978
1003
|
purpose=args.purpose, context=assignment_context, provider=args.provider,
|
|
979
1004
|
model_role=args.model_role, model=args.model, instruction_path=instruction_path,
|
|
980
1005
|
prompt_path=prompt_path, requested_runner=args.execution_runner,
|
|
1006
|
+
review_inputs=review_inputs,
|
|
981
1007
|
))
|
|
@@ -7,6 +7,7 @@ from dataclasses import dataclass
|
|
|
7
7
|
from pathlib import Path
|
|
8
8
|
|
|
9
9
|
from ..assignment_resolver import AssignmentContext, resolve_dispatch_assignment
|
|
10
|
+
from ..code_review_invocation import CodeReviewInputs
|
|
10
11
|
from .invocation import (
|
|
11
12
|
AgentAudience,
|
|
12
13
|
AgentInstruction,
|
|
@@ -33,6 +34,7 @@ class StandaloneInvocationRequest:
|
|
|
33
34
|
instruction_path: Path
|
|
34
35
|
prompt_path: Path
|
|
35
36
|
requested_runner: str | None = None
|
|
37
|
+
review_inputs: CodeReviewInputs | None = None
|
|
36
38
|
|
|
37
39
|
|
|
38
40
|
def prepare_standalone_invocation(
|
|
@@ -83,6 +85,7 @@ def prepare_standalone_invocation(
|
|
|
83
85
|
duty_id=None,
|
|
84
86
|
invocation_ref=None,
|
|
85
87
|
attempt=1,
|
|
88
|
+
review_inputs=request.review_inputs,
|
|
86
89
|
)
|
|
87
90
|
)
|
|
88
91
|
|
|
@@ -5,6 +5,8 @@ import hashlib
|
|
|
5
5
|
import re
|
|
6
6
|
from pathlib import Path
|
|
7
7
|
|
|
8
|
+
from okstra_project.dirs import contained_relative_path as project_relative_ref
|
|
9
|
+
|
|
8
10
|
from . import group_context
|
|
9
11
|
|
|
10
12
|
|
|
@@ -575,25 +577,6 @@ def exceeds_cited_document_limit(size: int) -> bool:
|
|
|
575
577
|
return size > MAX_CITED_DOCUMENT_BYTES
|
|
576
578
|
|
|
577
579
|
|
|
578
|
-
def project_relative_ref(ref: str, root: Path) -> str | None:
|
|
579
|
-
"""The posix path of a Source Material `ref` relative to the resolved `root`.
|
|
580
|
-
|
|
581
|
-
None when the ref resolves outside the root, or names the home directory of
|
|
582
|
-
a user this machine does not know (`~200` in prose). Every reader and
|
|
583
|
-
writer of a `### Source N — file` ref goes through here so they agree on
|
|
584
|
-
which file it names.
|
|
585
|
-
"""
|
|
586
|
-
try:
|
|
587
|
-
candidate = Path(ref).expanduser()
|
|
588
|
-
except RuntimeError:
|
|
589
|
-
return None
|
|
590
|
-
path = (candidate if candidate.is_absolute() else root / candidate).resolve()
|
|
591
|
-
try:
|
|
592
|
-
return path.relative_to(root).as_posix()
|
|
593
|
-
except ValueError:
|
|
594
|
-
return None
|
|
595
|
-
|
|
596
|
-
|
|
597
580
|
def _by_reference_file_refs(source_material: str) -> list[str]:
|
|
598
581
|
refs = [ref for ref, fmt in file_source_entries(source_material) if fmt == BY_REFERENCE_FORMAT]
|
|
599
582
|
return list(dict.fromkeys(refs))
|
|
@@ -164,6 +164,16 @@ def _stream_payload(path: Path) -> dict[str, Any]:
|
|
|
164
164
|
}
|
|
165
165
|
|
|
166
166
|
|
|
167
|
+
def read_sealed_artifact(seal: AttemptEvidenceSeal, path: Path) -> bytes:
|
|
168
|
+
row = next((row for row in seal.artifacts if row.get("normalizedPath") == str(path)), None)
|
|
169
|
+
if row is None or path.is_symlink() or not path.is_file():
|
|
170
|
+
raise AttemptEvidenceError(f"artifact is absent from the attempt seal: {path}")
|
|
171
|
+
data = path.read_bytes()
|
|
172
|
+
if len(data) != row.get("byteLength") or _sha(data) != row.get("sha256"):
|
|
173
|
+
raise AttemptEvidenceError(f"sealed artifact bytes changed: {path}")
|
|
174
|
+
return data
|
|
175
|
+
|
|
176
|
+
|
|
167
177
|
def _build_seal(
|
|
168
178
|
attempt_ref: str, terminal: TerminalObservation
|
|
169
179
|
) -> AttemptEvidenceSeal:
|
|
@@ -2,12 +2,14 @@
|
|
|
2
2
|
from __future__ import annotations
|
|
3
3
|
|
|
4
4
|
import ast
|
|
5
|
+
import json
|
|
5
6
|
import re
|
|
6
7
|
import subprocess
|
|
7
8
|
from dataclasses import dataclass
|
|
8
9
|
from pathlib import Path, PurePosixPath
|
|
10
|
+
from collections.abc import Mapping
|
|
9
11
|
|
|
10
|
-
from .
|
|
12
|
+
from .review_verdicts import read_review_census
|
|
11
13
|
|
|
12
14
|
_GROUPS = ("structural", "semantic", "state-and-tests", "general")
|
|
13
15
|
_FILE_GROUPS = {
|
|
@@ -20,7 +22,7 @@ _LOCKFILES = {
|
|
|
20
22
|
"Cargo.lock", "uv.lock", "Gemfile.lock", "composer.lock",
|
|
21
23
|
}
|
|
22
24
|
_GENERATED_ROOTS = ("dist/", "runtime/", "build/")
|
|
23
|
-
_DATA_SUFFIXES = {".
|
|
25
|
+
_DATA_SUFFIXES = {".csv"}
|
|
24
26
|
_PROSE_SUFFIXES = {".md", ".markdown", ".rst", ".txt"}
|
|
25
27
|
_UNPARSED_CODE_SUFFIXES = {
|
|
26
28
|
".js", ".mjs", ".cjs", ".ts", ".mts", ".cts", ".tsx", ".jsx", ".sh", ".bash",
|
|
@@ -43,8 +45,9 @@ class CensusResult:
|
|
|
43
45
|
def build_review_census(
|
|
44
46
|
repo: Path, base: str, head: str, rules_source: Path,
|
|
45
47
|
changed: dict[str, set[int]], *, previous: Path | None = None,
|
|
48
|
+
frozen_rules: str | None = None, inventory: Mapping[str, object] | None = None,
|
|
46
49
|
) -> CensusResult:
|
|
47
|
-
rules = _rules_section(rules_source
|
|
50
|
+
rules = _rules_section(rules_source) if frozen_rules is None else frozen_rules
|
|
48
51
|
files = _changed_files(repo, base, head)
|
|
49
52
|
cells: list[tuple[str, str]] = []
|
|
50
53
|
exclusions: list[tuple[str, str]] = []
|
|
@@ -57,13 +60,23 @@ def build_review_census(
|
|
|
57
60
|
continue
|
|
58
61
|
kind = "test" if _TEST_PATH.search(path) else "source"
|
|
59
62
|
cells.extend((path, group) for group in _FILE_GROUPS[kind])
|
|
60
|
-
if
|
|
63
|
+
if path not in changed:
|
|
61
64
|
continue
|
|
62
65
|
suffix = PurePosixPath(path).suffix
|
|
63
66
|
if suffix == ".py":
|
|
64
67
|
functions.extend(_changed_functions(repo, head, path, changed[path]))
|
|
65
68
|
elif suffix in _UNPARSED_CODE_SUFFIXES:
|
|
66
|
-
|
|
69
|
+
if inventory is None or path not in inventory:
|
|
70
|
+
unresolved.append(path)
|
|
71
|
+
else:
|
|
72
|
+
functions.extend(_supplement_functions(repo, head, path, changed[path],
|
|
73
|
+
inventory[path]))
|
|
74
|
+
if inventory is not None:
|
|
75
|
+
eligible = {path for path, _ in files if path in changed
|
|
76
|
+
and PurePosixPath(path).suffix in _UNPARSED_CODE_SUFFIXES
|
|
77
|
+
and _exclusion(path, False) is None}
|
|
78
|
+
if inventory.keys() - eligible:
|
|
79
|
+
raise ValueError("function inventory names files outside the unresolved census")
|
|
67
80
|
cells.extend((target, "semantic") for target in functions)
|
|
68
81
|
ordered = sorted(cells, key=lambda cell: _GROUPS.index(cell[1]))
|
|
69
82
|
ids, retired, new = _assign_ids(ordered, previous)
|
|
@@ -72,7 +85,10 @@ def build_review_census(
|
|
|
72
85
|
return CensusResult(text, tuple(unresolved))
|
|
73
86
|
|
|
74
87
|
|
|
75
|
-
def _rules_section(
|
|
88
|
+
def _rules_section(source: Path) -> str:
|
|
89
|
+
text = source.read_text(encoding="utf-8")
|
|
90
|
+
if source.suffix == ".json":
|
|
91
|
+
return render_rule_catalog(text, source.parent)
|
|
76
92
|
start = text.find("## Rules")
|
|
77
93
|
if start < 0:
|
|
78
94
|
raise ValueError("rules source has no ## Rules table")
|
|
@@ -80,6 +96,52 @@ def _rules_section(text: str) -> str:
|
|
|
80
96
|
return text[start:] if end < 0 else text[start:end + 1]
|
|
81
97
|
|
|
82
98
|
|
|
99
|
+
def render_rule_catalog(text: str, parent: Path, *, include_architecture: bool = True) -> str:
|
|
100
|
+
catalog = json.loads(text)
|
|
101
|
+
if (not isinstance(catalog, dict) or catalog.get("schemaVersion") != "1.0"
|
|
102
|
+
or not isinstance(catalog.get("rules"), dict)):
|
|
103
|
+
raise ValueError("unsupported review rule catalog")
|
|
104
|
+
rows = ["## Rules", "| Rule | Group | Reference |", "|---|---|---|"]
|
|
105
|
+
for identity, definition in catalog["rules"].items():
|
|
106
|
+
if (not isinstance(definition, list) or len(definition) != 2
|
|
107
|
+
or not all(isinstance(value, str) for value in definition)):
|
|
108
|
+
raise ValueError(f"invalid review rule: {identity}")
|
|
109
|
+
group, reference = definition
|
|
110
|
+
if not include_architecture and identity in {"S5", "S6", "S7", "S8", "S9", "S10"}:
|
|
111
|
+
continue
|
|
112
|
+
path, _, heading = reference.partition("#")
|
|
113
|
+
rows.append(f"| {identity} | {group} | {parent / path}#{heading} |")
|
|
114
|
+
return "\n".join(rows)
|
|
115
|
+
|
|
116
|
+
|
|
117
|
+
def _supplement_functions(
|
|
118
|
+
repo: Path, head: str, path: str, lines: set[int], entry: object,
|
|
119
|
+
) -> list[str]:
|
|
120
|
+
if (not isinstance(entry, dict) or set(entry) != {"functions", "rationale"}
|
|
121
|
+
or not isinstance(entry["rationale"], str) or not entry["rationale"].strip()
|
|
122
|
+
or not isinstance(entry["functions"], list)):
|
|
123
|
+
raise ValueError(f"function inventory needs functions and a completeness rationale: {path}")
|
|
124
|
+
source = subprocess.run(["git", "-C", str(repo), "show", f"{head}:{path}"],
|
|
125
|
+
capture_output=True, text=True, check=True).stdout.splitlines()
|
|
126
|
+
targets: list[str] = []
|
|
127
|
+
seen: set[tuple[int, int, str]] = set()
|
|
128
|
+
for row in entry["functions"]:
|
|
129
|
+
if not isinstance(row, dict) or set(row) != {"start", "end", "name"}:
|
|
130
|
+
raise ValueError(f"invalid function inventory row: {path}")
|
|
131
|
+
start, end, name = row["start"], row["end"], row["name"]
|
|
132
|
+
if (type(start) is not int or type(end) is not int
|
|
133
|
+
or not 1 <= start <= end <= len(source)
|
|
134
|
+
or not isinstance(name, str) or not re.fullmatch(r"[\w.$<>:-]+", name)):
|
|
135
|
+
raise ValueError(f"invalid function inventory span or name: {path}")
|
|
136
|
+
identity = (start, end, name)
|
|
137
|
+
if identity in seen:
|
|
138
|
+
raise ValueError(f"duplicate function inventory row: {path}")
|
|
139
|
+
seen.add(identity)
|
|
140
|
+
if lines.intersection(range(start, end + 1)):
|
|
141
|
+
targets.append(f"{path}:{start} {name}")
|
|
142
|
+
return targets
|
|
143
|
+
|
|
144
|
+
|
|
83
145
|
def _changed_files(repo: Path, base: str, head: str) -> list[tuple[str, bool]]:
|
|
84
146
|
listed = subprocess.run(
|
|
85
147
|
["git", "-C", str(repo), "diff", "--numstat", "-z", "-M", "--no-ext-diff",
|
|
@@ -111,8 +173,10 @@ def _exclusion(path: str, binary: bool) -> str | None:
|
|
|
111
173
|
return "generated build output"
|
|
112
174
|
if binary:
|
|
113
175
|
return "binary file"
|
|
176
|
+
if name in ("AGENTS.md", "CLAUDE.md"):
|
|
177
|
+
return None
|
|
114
178
|
if suffix in _DATA_SUFFIXES:
|
|
115
|
-
return "
|
|
179
|
+
return "data without executable project code"
|
|
116
180
|
if suffix in _PROSE_SUFFIXES:
|
|
117
181
|
return "prose documentation without executable or agent-directive content"
|
|
118
182
|
return None
|