@agent-delivery-harness/conformance 0.1.0
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/LICENSE +202 -0
- package/NOTICE +5 -0
- package/README.md +159 -0
- package/fixtures/repo-config-adapter.ts +250 -0
- package/package.json +24 -0
- package/src/generate.ts +1013 -0
- package/src/index.ts +36 -0
- package/src/json-schema.ts +222 -0
- package/src/runner.ts +687 -0
- package/vectors/context/environment.json +21 -0
- package/vectors/context/repo-config.json +22 -0
- package/vectors/kit.json +1250 -0
- package/vectors/vectors/accept/a-completed-order-differs.json +112 -0
- package/vectors/vectors/accept/a-cost-fractional.json +116 -0
- package/vectors/vectors/accept/a-cost-tokens.json +120 -0
- package/vectors/vectors/accept/a-deferred-expansion.json +124 -0
- package/vectors/vectors/accept/a-idempotent-resubmission.json +115 -0
- package/vectors/vectors/accept/a-minimal-green.json +97 -0
- package/vectors/vectors/accept/a-resolved-and-advisory.json +129 -0
- package/vectors/vectors/accept/a-three-reviewers.json +113 -0
- package/vectors/vectors/reject/env-1-unregistered-provider.json +114 -0
- package/vectors/vectors/reject/env-10-artifact-duplicate.json +103 -0
- package/vectors/vectors/reject/env-10-artifact-missing-file.json +104 -0
- package/vectors/vectors/reject/env-10-artifact-traversal.json +99 -0
- package/vectors/vectors/reject/env-11-artifact-digest-mismatch.json +98 -0
- package/vectors/vectors/reject/env-12-unknown-attestation-level.json +114 -0
- package/vectors/vectors/reject/env-13-nonempty-signatures.json +118 -0
- package/vectors/vectors/reject/env-14-duplicate-claim.json +149 -0
- package/vectors/vectors/reject/env-14-no-claims.json +78 -0
- package/vectors/vectors/reject/env-14-unconfigured-obligation.json +114 -0
- package/vectors/vectors/reject/env-14-unsupported-payload-spec.json +114 -0
- package/vectors/vectors/reject/env-2-run-id-1.json +114 -0
- package/vectors/vectors/reject/env-2-run-id-2.json +114 -0
- package/vectors/vectors/reject/env-2-run-id-3.json +114 -0
- package/vectors/vectors/reject/env-2-run-id-4.json +114 -0
- package/vectors/vectors/reject/env-2-run-id-5.json +114 -0
- package/vectors/vectors/reject/env-4-unsupported-vcs.json +114 -0
- package/vectors/vectors/reject/env-5-malformed-object-id.json +114 -0
- package/vectors/vectors/reject/env-6-missing-identity.json +111 -0
- package/vectors/vectors/reject/env-6-unknown-identity-version.json +114 -0
- package/vectors/vectors/reject/env-8-repository-required.json +116 -0
- package/vectors/vectors/reject/env-9-empty-run-history.json +107 -0
- package/vectors/vectors/reject/env-9-final-pass-mismatch.json +110 -0
- package/vectors/vectors/reject/env-9-final-tree-mismatch.json +110 -0
- package/vectors/vectors/reject/gen-1-legacy-count-field.json +115 -0
- package/vectors/vectors/reject/gen-1-unknown-envelope-member.json +117 -0
- package/vectors/vectors/reject/gen-2-unsupported-spec.json +114 -0
- package/vectors/vectors/reject/rg-1-edited-after-final-pass.json +114 -0
- package/vectors/vectors/reject/rg-1-not-finalized.json +114 -0
- package/vectors/vectors/reject/rg-1-verdict-not-green.json +114 -0
- package/vectors/vectors/reject/rg-10-cost-by-reviewer-exceeds-total.json +115 -0
- package/vectors/vectors/reject/rg-10-cost-empty-reported-by.json +111 -0
- package/vectors/vectors/reject/rg-10-cost-empty-unit.json +111 -0
- package/vectors/vectors/reject/rg-10-cost-missing-unit.json +110 -0
- package/vectors/vectors/reject/rg-10-cost-negative-by-reviewer.json +114 -0
- package/vectors/vectors/reject/rg-10-cost-negative-total.json +111 -0
- package/vectors/vectors/reject/rg-10-cost-null-total.json +111 -0
- package/vectors/vectors/reject/rg-10-cost-unknown-reviewer-in-breakdown.json +114 -0
- package/vectors/vectors/reject/rg-2-duplicate-selected.json +99 -0
- package/vectors/vectors/reject/rg-3-completed-missing.json +96 -0
- package/vectors/vectors/reject/rg-3-completed-unrelated.json +99 -0
- package/vectors/vectors/reject/rg-3-failed-reviewer.json +100 -0
- package/vectors/vectors/reject/rg-3-timed-out-reviewer.json +100 -0
- package/vectors/vectors/reject/rg-4-approval-extra.json +104 -0
- package/vectors/vectors/reject/rg-4-approval-malformed.json +98 -0
- package/vectors/vectors/reject/rg-4-approval-missing.json +100 -0
- package/vectors/vectors/reject/rg-4-approval-not-approved.json +98 -0
- package/vectors/vectors/reject/rg-4-approval-wrong-candidate.json +98 -0
- package/vectors/vectors/reject/rg-4-approval-wrong-pass.json +98 -0
- package/vectors/vectors/reject/rg-4-approval-wrong-provider.json +98 -0
- package/vectors/vectors/reject/rg-4-approval-wrong-reviewer.json +98 -0
- package/vectors/vectors/reject/rg-4-approval-wrong-run.json +98 -0
- package/vectors/vectors/reject/rg-4-approval-wrong-workspace.json +98 -0
- package/vectors/vectors/reject/rg-5-duplicate-finding-id.json +131 -0
- package/vectors/vectors/reject/rg-5-unknown-disposition.json +123 -0
- package/vectors/vectors/reject/rg-5-unknown-scope.json +123 -0
- package/vectors/vectors/reject/rg-5-unknown-severity.json +123 -0
- package/vectors/vectors/reject/rg-6-actionable-ignored.json +123 -0
- package/vectors/vectors/reject/rg-6-actionable-unresolved.json +123 -0
- package/vectors/vectors/reject/rg-6-blocking-finding.json +123 -0
- package/vectors/vectors/reject/rg-7-defer-adjacent.json +126 -0
- package/vectors/vectors/reject/rg-7-defer-blocking.json +126 -0
- package/vectors/vectors/reject/rg-7-defer-in-contract.json +126 -0
- package/vectors/vectors/reject/rg-7-defer-lowercase-issue-id.json +126 -0
- package/vectors/vectors/reject/rg-7-defer-missing-issue-id.json +125 -0
- package/vectors/vectors/reject/rg-7-defer-non-actionable.json +126 -0
- package/vectors/vectors/reject/rg-7-defer-p0.json +126 -0
- package/vectors/vectors/reject/rg-7-defer-p1.json +126 -0
- package/vectors/vectors/reject/rg-7-defer-placeholder-issue-id.json +126 -0
- package/vectors/vectors/reject/rg-7-issue-id-on-resolved.json +124 -0
- package/vectors/vectors/reject/rg-8-counts-contradict-findings.json +126 -0
- package/vectors/vectors/reject/rg-8-deferral-count-disagrees.json +124 -0
- package/vectors/vectors/reject/rg-8-deferral-ids-disagree.json +126 -0
- package/vectors/vectors/reject/rg-9-iteration-count-mismatch.json +114 -0
- package/vectors/vectors/reject/sub-1-base-tip-moved.json +131 -0
- package/vectors/vectors/reject/sub-1-deliverable-changed.json +131 -0
- package/vectors/vectors/reject/sub-1-merge-base-moved.json +131 -0
- package/vectors/vectors/reject/sub-1-raw-tree-changed.json +131 -0
- package/vectors/vectors/reject/sub-2-unprepared.json +117 -0
- package/vectors/vectors/reject/sub-3-manifest-outside-run-root.json +117 -0
- package/vectors/vectors/reject/sub-4-record-conflict.json +220 -0
package/LICENSE
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
1
|
+
|
|
2
|
+
Apache License
|
|
3
|
+
Version 2.0, January 2004
|
|
4
|
+
http://www.apache.org/licenses/
|
|
5
|
+
|
|
6
|
+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
|
7
|
+
|
|
8
|
+
1. Definitions.
|
|
9
|
+
|
|
10
|
+
"License" shall mean the terms and conditions for use, reproduction,
|
|
11
|
+
and distribution as defined by Sections 1 through 9 of this document.
|
|
12
|
+
|
|
13
|
+
"Licensor" shall mean the copyright owner or entity authorized by
|
|
14
|
+
the copyright owner that is granting the License.
|
|
15
|
+
|
|
16
|
+
"Legal Entity" shall mean the union of the acting entity and all
|
|
17
|
+
other entities that control, are controlled by, or are under common
|
|
18
|
+
control with that entity. For the purposes of this definition,
|
|
19
|
+
"control" means (i) the power, direct or indirect, to cause the
|
|
20
|
+
direction or management of such entity, whether by contract or
|
|
21
|
+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
|
22
|
+
outstanding shares, or (iii) beneficial ownership of such entity.
|
|
23
|
+
|
|
24
|
+
"You" (or "Your") shall mean an individual or Legal Entity
|
|
25
|
+
exercising permissions granted by this License.
|
|
26
|
+
|
|
27
|
+
"Source" form shall mean the preferred form for making modifications,
|
|
28
|
+
including but not limited to software source code, documentation
|
|
29
|
+
source, and configuration files.
|
|
30
|
+
|
|
31
|
+
"Object" form shall mean any form resulting from mechanical
|
|
32
|
+
transformation or translation of a Source form, including but
|
|
33
|
+
not limited to compiled object code, generated documentation,
|
|
34
|
+
and conversions to other media types.
|
|
35
|
+
|
|
36
|
+
"Work" shall mean the work of authorship, whether in Source or
|
|
37
|
+
Object form, made available under the License, as indicated by a
|
|
38
|
+
copyright notice that is included in or attached to the work
|
|
39
|
+
(an example is provided in the Appendix below).
|
|
40
|
+
|
|
41
|
+
"Derivative Works" shall mean any work, whether in Source or Object
|
|
42
|
+
form, that is based on (or derived from) the Work and for which the
|
|
43
|
+
editorial revisions, annotations, elaborations, or other modifications
|
|
44
|
+
represent, as a whole, an original work of authorship. For the purposes
|
|
45
|
+
of this License, Derivative Works shall not include works that remain
|
|
46
|
+
separable from, or merely link (or bind by name) to the interfaces of,
|
|
47
|
+
the Work and Derivative Works thereof.
|
|
48
|
+
|
|
49
|
+
"Contribution" shall mean any work of authorship, including
|
|
50
|
+
the original version of the Work and any modifications or additions
|
|
51
|
+
to that Work or Derivative Works thereof, that is intentionally
|
|
52
|
+
submitted to Licensor for inclusion in the Work by the copyright owner
|
|
53
|
+
or by an individual or Legal Entity authorized to submit on behalf of
|
|
54
|
+
the copyright owner. For the purposes of this definition, "submitted"
|
|
55
|
+
means any form of electronic, verbal, or written communication sent
|
|
56
|
+
to the Licensor or its representatives, including but not limited to
|
|
57
|
+
communication on electronic mailing lists, source code control systems,
|
|
58
|
+
and issue tracking systems that are managed by, or on behalf of, the
|
|
59
|
+
Licensor for the purpose of discussing and improving the Work, but
|
|
60
|
+
excluding communication that is conspicuously marked or otherwise
|
|
61
|
+
designated in writing by the copyright owner as "Not a Contribution."
|
|
62
|
+
|
|
63
|
+
"Contributor" shall mean Licensor and any individual or Legal Entity
|
|
64
|
+
on behalf of whom a Contribution has been received by Licensor and
|
|
65
|
+
subsequently incorporated within the Work.
|
|
66
|
+
|
|
67
|
+
2. Grant of Copyright License. Subject to the terms and conditions of
|
|
68
|
+
this License, each Contributor hereby grants to You a perpetual,
|
|
69
|
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
|
70
|
+
copyright license to reproduce, prepare Derivative Works of,
|
|
71
|
+
publicly display, publicly perform, sublicense, and distribute the
|
|
72
|
+
Work and such Derivative Works in Source or Object form.
|
|
73
|
+
|
|
74
|
+
3. Grant of Patent License. Subject to the terms and conditions of
|
|
75
|
+
this License, each Contributor hereby grants to You a perpetual,
|
|
76
|
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
|
77
|
+
(except as stated in this section) patent license to make, have made,
|
|
78
|
+
use, offer to sell, sell, import, and otherwise transfer the Work,
|
|
79
|
+
where such license applies only to those patent claims licensable
|
|
80
|
+
by such Contributor that are necessarily infringed by their
|
|
81
|
+
Contribution(s) alone or by combination of their Contribution(s)
|
|
82
|
+
with the Work to which such Contribution(s) was submitted. If You
|
|
83
|
+
institute patent litigation against any entity (including a
|
|
84
|
+
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
|
85
|
+
or a Contribution incorporated within the Work constitutes direct
|
|
86
|
+
or contributory patent infringement, then any patent licenses
|
|
87
|
+
granted to You under this License for that Work shall terminate
|
|
88
|
+
as of the date such litigation is filed.
|
|
89
|
+
|
|
90
|
+
4. Redistribution. You may reproduce and distribute copies of the
|
|
91
|
+
Work or Derivative Works thereof in any medium, with or without
|
|
92
|
+
modifications, and in Source or Object form, provided that You
|
|
93
|
+
meet the following conditions:
|
|
94
|
+
|
|
95
|
+
(a) You must give any other recipients of the Work or
|
|
96
|
+
Derivative Works a copy of this License; and
|
|
97
|
+
|
|
98
|
+
(b) You must cause any modified files to carry prominent notices
|
|
99
|
+
stating that You changed the files; and
|
|
100
|
+
|
|
101
|
+
(c) You must retain, in the Source form of any Derivative Works
|
|
102
|
+
that You distribute, all copyright, patent, trademark, and
|
|
103
|
+
attribution notices from the Source form of the Work,
|
|
104
|
+
excluding those notices that do not pertain to any part of
|
|
105
|
+
the Derivative Works; and
|
|
106
|
+
|
|
107
|
+
(d) If the Work includes a "NOTICE" text file as part of its
|
|
108
|
+
distribution, then any Derivative Works that You distribute must
|
|
109
|
+
include a readable copy of the attribution notices contained
|
|
110
|
+
within such NOTICE file, excluding those notices that do not
|
|
111
|
+
pertain to any part of the Derivative Works, in at least one
|
|
112
|
+
of the following places: within a NOTICE text file distributed
|
|
113
|
+
as part of the Derivative Works; within the Source form or
|
|
114
|
+
documentation, if provided along with the Derivative Works; or,
|
|
115
|
+
within a display generated by the Derivative Works, if and
|
|
116
|
+
wherever such third-party notices normally appear. The contents
|
|
117
|
+
of the NOTICE file are for informational purposes only and
|
|
118
|
+
do not modify the License. You may add Your own attribution
|
|
119
|
+
notices within Derivative Works that You distribute, alongside
|
|
120
|
+
or as an addendum to the NOTICE text from the Work, provided
|
|
121
|
+
that such additional attribution notices cannot be construed
|
|
122
|
+
as modifying the License.
|
|
123
|
+
|
|
124
|
+
You may add Your own copyright statement to Your modifications and
|
|
125
|
+
may provide additional or different license terms and conditions
|
|
126
|
+
for use, reproduction, or distribution of Your modifications, or
|
|
127
|
+
for any such Derivative Works as a whole, provided Your use,
|
|
128
|
+
reproduction, and distribution of the Work otherwise complies with
|
|
129
|
+
the conditions stated in this License.
|
|
130
|
+
|
|
131
|
+
5. Submission of Contributions. Unless You explicitly state otherwise,
|
|
132
|
+
any Contribution intentionally submitted for inclusion in the Work
|
|
133
|
+
by You to the Licensor shall be under the terms and conditions of
|
|
134
|
+
this License, without any additional terms or conditions.
|
|
135
|
+
Notwithstanding the above, nothing herein shall supersede or modify
|
|
136
|
+
the terms of any separate license agreement you may have executed
|
|
137
|
+
with Licensor regarding such Contributions.
|
|
138
|
+
|
|
139
|
+
6. Trademarks. This License does not grant permission to use the trade
|
|
140
|
+
names, trademarks, service marks, or product names of the Licensor,
|
|
141
|
+
except as required for reasonable and customary use in describing the
|
|
142
|
+
origin of the Work and reproducing the content of the NOTICE file.
|
|
143
|
+
|
|
144
|
+
7. Disclaimer of Warranty. Unless required by applicable law or
|
|
145
|
+
agreed to in writing, Licensor provides the Work (and each
|
|
146
|
+
Contributor provides its Contributions) on an "AS IS" BASIS,
|
|
147
|
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
|
148
|
+
implied, including, without limitation, any warranties or conditions
|
|
149
|
+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
|
150
|
+
PARTICULAR PURPOSE. You are solely responsible for determining the
|
|
151
|
+
appropriateness of using or redistributing the Work and assume any
|
|
152
|
+
risks associated with Your exercise of permissions under this License.
|
|
153
|
+
|
|
154
|
+
8. Limitation of Liability. In no event and under no legal theory,
|
|
155
|
+
whether in tort (including negligence), contract, or otherwise,
|
|
156
|
+
unless required by applicable law (such as deliberate and grossly
|
|
157
|
+
negligent acts) or agreed to in writing, shall any Contributor be
|
|
158
|
+
liable to You for damages, including any direct, indirect, special,
|
|
159
|
+
incidental, or consequential damages of any character arising as a
|
|
160
|
+
result of this License or out of the use or inability to use the
|
|
161
|
+
Work (including but not limited to damages for loss of goodwill,
|
|
162
|
+
work stoppage, computer failure or malfunction, or any and all
|
|
163
|
+
other commercial damages or losses), even if such Contributor
|
|
164
|
+
has been advised of the possibility of such damages.
|
|
165
|
+
|
|
166
|
+
9. Accepting Warranty or Additional Liability. While redistributing
|
|
167
|
+
the Work or Derivative Works thereof, You may choose to offer,
|
|
168
|
+
and charge a fee for, acceptance of support, warranty, indemnity,
|
|
169
|
+
or other liability obligations and/or rights consistent with this
|
|
170
|
+
License. However, in accepting such obligations, You may act only
|
|
171
|
+
on Your own behalf and on Your sole responsibility, not on behalf
|
|
172
|
+
of any other Contributor, and only if You agree to indemnify,
|
|
173
|
+
defend, and hold each Contributor harmless for any liability
|
|
174
|
+
incurred by, or claims asserted against, such Contributor by reason
|
|
175
|
+
of your accepting any such warranty or additional liability.
|
|
176
|
+
|
|
177
|
+
END OF TERMS AND CONDITIONS
|
|
178
|
+
|
|
179
|
+
APPENDIX: How to apply the Apache License to your work.
|
|
180
|
+
|
|
181
|
+
To apply the Apache License to your work, attach the following
|
|
182
|
+
boilerplate notice, with the fields enclosed by brackets "[]"
|
|
183
|
+
replaced with your own identifying information. (Don't include
|
|
184
|
+
the brackets!) The text should be enclosed in the appropriate
|
|
185
|
+
comment syntax for the file format. We also recommend that a
|
|
186
|
+
file or class name and description of purpose be included on the
|
|
187
|
+
same "printed page" as the copyright notice for easier
|
|
188
|
+
identification within third-party archives.
|
|
189
|
+
|
|
190
|
+
Copyright [yyyy] [name of copyright owner]
|
|
191
|
+
|
|
192
|
+
Licensed under the Apache License, Version 2.0 (the "License");
|
|
193
|
+
you may not use this file except in compliance with the License.
|
|
194
|
+
You may obtain a copy of the License at
|
|
195
|
+
|
|
196
|
+
http://www.apache.org/licenses/LICENSE-2.0
|
|
197
|
+
|
|
198
|
+
Unless required by applicable law or agreed to in writing, software
|
|
199
|
+
distributed under the License is distributed on an "AS IS" BASIS,
|
|
200
|
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
201
|
+
See the License for the specific language governing permissions and
|
|
202
|
+
limitations under the License.
|
package/NOTICE
ADDED
package/README.md
ADDED
|
@@ -0,0 +1,159 @@
|
|
|
1
|
+
# Delivery Evidence Conformance Kit
|
|
2
|
+
|
|
3
|
+
Golden vectors for validators of the **delivery-evidence/1** specification
|
|
4
|
+
(envelope) and the **review.green/1** payload. A validator implementation that
|
|
5
|
+
produces the expected outcome for every vector conforms to the machine-checkable
|
|
6
|
+
core of the spec.
|
|
7
|
+
|
|
8
|
+
The corpus is derived from a production recorder's test suite
|
|
9
|
+
(`athena:scripts/harness-review-evidence.test.ts`, ~1,000 lines) that has gated
|
|
10
|
+
agent-delivered merges since mid-2026, restated against the generalized
|
|
11
|
+
envelope. Provenance is recorded per vector.
|
|
12
|
+
|
|
13
|
+
## Layout
|
|
14
|
+
|
|
15
|
+
```
|
|
16
|
+
kit.json index: every vector with id, file, rules, expectation
|
|
17
|
+
context/repo-config.json the repository gate configuration vectors assume
|
|
18
|
+
context/environment.json the submission-time state the harness must establish
|
|
19
|
+
vectors/accept/*.json manifests a conforming validator MUST accept
|
|
20
|
+
vectors/reject/*.json manifests a conforming validator MUST reject
|
|
21
|
+
generate.ts regenerates everything above (bun generate.ts <outDir>)
|
|
22
|
+
```
|
|
23
|
+
|
|
24
|
+
## Vector format
|
|
25
|
+
|
|
26
|
+
Each vector is self-contained:
|
|
27
|
+
|
|
28
|
+
```jsonc
|
|
29
|
+
{
|
|
30
|
+
"vectorVersion": 1,
|
|
31
|
+
"id": "rg-7-defer-p0",
|
|
32
|
+
"title": "Illegal deferral: p0",
|
|
33
|
+
"rules": ["RG-7"], // spec rule ids this vector exercises
|
|
34
|
+
"provenance": "athena:… | new in spec | spec-tightened",
|
|
35
|
+
"expect": { "result": "rejected", "codes": ["illegal_deferral"] },
|
|
36
|
+
"environment": { … }, // optional overrides of context/environment.json
|
|
37
|
+
"extra": { … }, // optional protocol notes (multi-step vectors)
|
|
38
|
+
"artifacts": { "reviewers/correctness.json": "<exact file bytes>" },
|
|
39
|
+
"manifest": { … } // the delivery-evidence/1 manifest to submit
|
|
40
|
+
}
|
|
41
|
+
```
|
|
42
|
+
|
|
43
|
+
## Running a vector
|
|
44
|
+
|
|
45
|
+
For each vector, the harness under test:
|
|
46
|
+
|
|
47
|
+
1. Loads `context/repo-config.json` as the repository's gate configuration.
|
|
48
|
+
2. Establishes the submission environment from `context/environment.json`,
|
|
49
|
+
merged with the vector's `environment` overrides:
|
|
50
|
+
- `currentCandidate` is what the recorder's candidate capture returns at
|
|
51
|
+
submission time.
|
|
52
|
+
- `prepared: false` means preparation is absent/failed (the capture reports
|
|
53
|
+
an unprepared state).
|
|
54
|
+
- `manifestLocation: "outside-run-root"` means the harness places the
|
|
55
|
+
manifest file somewhere other than the allocated run root.
|
|
56
|
+
3. Allocates a run root for `manifest.provider.runId` and materializes each
|
|
57
|
+
entry of `artifacts` at its path (exact bytes) relative to that root. An
|
|
58
|
+
artifact path that is itself invalid (e.g. `../outside.json`) is part of the
|
|
59
|
+
vector's point; materialize it where the path indicates, or skip
|
|
60
|
+
materialization if the validator rejects on path shape before file access.
|
|
61
|
+
4. Submits the manifest through the recorder.
|
|
62
|
+
5. Compares the outcome to `expect`.
|
|
63
|
+
|
|
64
|
+
### Multi-step vectors
|
|
65
|
+
|
|
66
|
+
- `a-idempotent-resubmission` (`extra.submitTwice`): submit the same manifest
|
|
67
|
+
twice. Both submissions succeed and the record ids are identical.
|
|
68
|
+
- `sub-4-record-conflict` (`extra.submitFirst`): submit `extra.submitFirst`
|
|
69
|
+
(accepted), then the vector's `manifest`. Same record identity, different
|
|
70
|
+
content — the second submission is rejected with `record_conflict`.
|
|
71
|
+
|
|
72
|
+
## Expectation semantics
|
|
73
|
+
|
|
74
|
+
- `result: "accepted"` — the submission MUST be accepted and one evidence
|
|
75
|
+
record written per claim. No rejection codes.
|
|
76
|
+
- `result: "rejected"` — the submission MUST be rejected, no record written,
|
|
77
|
+
and the response MUST include **every** code listed in `expect.codes`.
|
|
78
|
+
Additional codes are permitted: several vectors violate more than one rule by
|
|
79
|
+
construction (SUB-5 requires reporting all violations), and the kit does not
|
|
80
|
+
over-specify cascade behavior. The listed codes are the floor, not the
|
|
81
|
+
ceiling.
|
|
82
|
+
|
|
83
|
+
## Provenance markers
|
|
84
|
+
|
|
85
|
+
- `athena:…` — restates a case from the production test corpus.
|
|
86
|
+
- `new in spec; no Athena antecedent` — covers spec surface the production
|
|
87
|
+
contract did not have (claims array, artifact digests, attestation levels,
|
|
88
|
+
strict unknown-member parsing, repository requirement).
|
|
89
|
+
- `spec-tightened; Athena accepted a looser form` — the spec is deliberately
|
|
90
|
+
stricter here; the production recorder accepted this shape. Two cases:
|
|
91
|
+
- `env-9-empty-run-history` — Athena accepted an empty mutation sequence;
|
|
92
|
+
the spec requires a non-empty `runHistory` whose final entry names the
|
|
93
|
+
candidate tree and final pass.
|
|
94
|
+
- `rg-9-iteration-count-mismatch` — Athena accepted an explicit
|
|
95
|
+
`iterationCount` that disagreed with the mutation sequence; the spec
|
|
96
|
+
requires `iterationCount == runHistory.length`.
|
|
97
|
+
|
|
98
|
+
## Known divergences from the production contract
|
|
99
|
+
|
|
100
|
+
Beyond the two tightenings above, the spec-shaped manifest differs from the
|
|
101
|
+
production manifest in ways the kit reflects:
|
|
102
|
+
|
|
103
|
+
1. **Claims array.** Production bound one manifest to one obligation; the
|
|
104
|
+
envelope carries `claims[]` and the recorder writes one record per claim.
|
|
105
|
+
2. **Artifact digests.** Production validated artifacts at record time but did
|
|
106
|
+
not hash them into the manifest; the envelope requires per-artifact
|
|
107
|
+
`sha256` (`env-11-artifact-digest-mismatch` is new surface).
|
|
108
|
+
3. **Strict parsing.** Production tolerated some unknown members; the spec
|
|
109
|
+
rejects any (`gen-1-*`). Production's summary-count fields
|
|
110
|
+
(`unresolvedActionableCount`, `blockingCount`) were removed as pure
|
|
111
|
+
derivations — their presence now trips `unknown_member`
|
|
112
|
+
(`gen-1-legacy-count-field`).
|
|
113
|
+
4. **40-hex object ids.** Production fixtures used symbolic ids
|
|
114
|
+
(`"tree-a"`); the spec requires real git object-id shapes, so all kit ids
|
|
115
|
+
are deterministic hex derived in `generate.ts`.
|
|
116
|
+
5. **Approval stamp shape.** Production stamps carried flat
|
|
117
|
+
`providerId`/`runId` fields; the spec nests them under `provider` and
|
|
118
|
+
renames `worktreeId` to `workspaceId` (§9.2).
|
|
119
|
+
|
|
120
|
+
## Coverage
|
|
121
|
+
|
|
122
|
+
89 vectors: 8 accept, 81 reject. Every rule in spec §8 (GEN, ENV, SUB) and §9.3
|
|
123
|
+
(RG) is exercised by at least one vector except: GEN-3 and SUB-5 (meta rules
|
|
124
|
+
about validator behavior, observable through the multi-code vectors), GEN-5
|
|
125
|
+
(prohibition on time-based decisions — not observable from a single
|
|
126
|
+
submission), and ENV-3/ENV-7 (subsumed by schema-shape vectors). The signing
|
|
127
|
+
profile, additional payload specs, and gate-policy behavior (freshness at gate
|
|
128
|
+
time, waivers, delegation) are out of the kit's scope, as they are out of the
|
|
129
|
+
spec's.
|
|
130
|
+
|
|
131
|
+
## Regenerating
|
|
132
|
+
|
|
133
|
+
```
|
|
134
|
+
bun generate.ts <outDir>
|
|
135
|
+
```
|
|
136
|
+
|
|
137
|
+
The generator is table-driven — one entry per vector, mirroring the structure
|
|
138
|
+
of the source test corpus. Accept-vector artifact digests are computed from the
|
|
139
|
+
emitted bytes, so edits to artifact content stay internally consistent
|
|
140
|
+
automatically.
|
|
141
|
+
|
|
142
|
+
---
|
|
143
|
+
|
|
144
|
+
## Vendored layout in this repo
|
|
145
|
+
|
|
146
|
+
This README is the kit's own, carried verbatim. Two paths differ from the layout
|
|
147
|
+
above because the kit is vendored into an npm workspace:
|
|
148
|
+
|
|
149
|
+
| Kit path | Path here |
|
|
150
|
+
|---|---|
|
|
151
|
+
| kit root (`kit.json`, `context/`, `vectors/`) | `packages/conformance/vectors/` |
|
|
152
|
+
| `generate.ts` | `packages/conformance/src/generate.ts` |
|
|
153
|
+
|
|
154
|
+
Regenerate with `npm run kit:generate` (the generator is also runnable directly
|
|
155
|
+
under either runtime: `node --import tsx packages/conformance/src/generate.ts <outDir>`
|
|
156
|
+
or `bun packages/conformance/src/generate.ts <outDir>`). The vendored tree is
|
|
157
|
+
guarded against drift by `packages/conformance/src/generate.test.ts`, which
|
|
158
|
+
regenerates into a temp directory and requires byte-identical output — so the
|
|
159
|
+
vectors are never hand-edited, only regenerated.
|
|
@@ -0,0 +1,250 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* The conformance kit's `context/repo-config.json`, adapted into a harness
|
|
3
|
+
* config.
|
|
4
|
+
*
|
|
5
|
+
* The kit declares the repository policy a validator needs to judge its
|
|
6
|
+
* vectors, and nothing else: a gate id, the envelope specs and identity
|
|
7
|
+
* versions it accepts, and one obligation with its payload specs, providers and
|
|
8
|
+
* minimum attestation level. A harness config is a much wider surface, so the
|
|
9
|
+
* adapter does three separable things, and the test beside it checks each one:
|
|
10
|
+
*
|
|
11
|
+
* MAPS the fields the kit declares. These are the dimensions the vectors are
|
|
12
|
+
* bound to — a vector that expects `unregistered_provider` is expecting it
|
|
13
|
+
* against *this* provider list — so they are carried across verbatim.
|
|
14
|
+
*
|
|
15
|
+
* STRIPS `configVersion`. It versions the kit's own file format and has no
|
|
16
|
+
* counterpart in a harness config; carrying it would fail the closed grammar.
|
|
17
|
+
*
|
|
18
|
+
* DEFAULTS every remaining dimension. Each default is either vector-
|
|
19
|
+
* independent (no vector's expectation can turn on it) or forced by a config
|
|
20
|
+
* invariant. There is exactly one of the latter, and it is named below.
|
|
21
|
+
*
|
|
22
|
+
* THE ONE CONSTRAINED DEFAULT. The kit declares `deliverable-tree/v1` as its
|
|
23
|
+
* identity version, and the config invariants bind that token to its narration
|
|
24
|
+
* set in both directions. So `reviewNeutral` is not free here: it must be the
|
|
25
|
+
* `deliverable-tree/v1` narration set or the config will not load. This is
|
|
26
|
+
* harmless for kit runs, which stub candidate capture with each vector's own
|
|
27
|
+
* values and never recompute an identity from a real tree — but it is a
|
|
28
|
+
* constraint rather than a free choice, and pretending otherwise would hide the
|
|
29
|
+
* one place the adapter's hands are tied.
|
|
30
|
+
*
|
|
31
|
+
* NEGATIVE SPACE. Four tokens must stay *absent* from this config, because four
|
|
32
|
+
* reject vectors expect a rejection that only happens while they are absent:
|
|
33
|
+
* the obligation `qa.exercised`, the payload spec `review.green/0`, the provider
|
|
34
|
+
* `unknown.provider`, and the identity version `deliverable-tree/v0`. A default
|
|
35
|
+
* that quietly widened the config to include one of them would turn a reject
|
|
36
|
+
* vector green without changing a single expectation. `negativeSpaceViolations`
|
|
37
|
+
* exists so that widening is caught rather than discovered.
|
|
38
|
+
*/
|
|
39
|
+
import { readFileSync } from "node:fs";
|
|
40
|
+
import path from "node:path";
|
|
41
|
+
import { fileURLToPath } from "node:url";
|
|
42
|
+
import {
|
|
43
|
+
DELIVERABLE_TREE_V1_NARRATION_SET,
|
|
44
|
+
defineHarnessConfig,
|
|
45
|
+
type HarnessConfig,
|
|
46
|
+
type HarnessConfigInput,
|
|
47
|
+
} from "@agent-delivery-harness/kernel";
|
|
48
|
+
|
|
49
|
+
/** The kit's own file format, as declared in `context/repo-config.json`. */
|
|
50
|
+
export interface KitRepoConfig {
|
|
51
|
+
readonly configVersion: number;
|
|
52
|
+
readonly gateId: string;
|
|
53
|
+
readonly acceptedEnvelopeSpecs: readonly string[];
|
|
54
|
+
readonly identityVersions: readonly string[];
|
|
55
|
+
readonly obligations: Readonly<
|
|
56
|
+
Record<
|
|
57
|
+
string,
|
|
58
|
+
{
|
|
59
|
+
readonly acceptedPayloadSpecs: readonly string[];
|
|
60
|
+
readonly providers: readonly string[];
|
|
61
|
+
readonly minimumAttestationLevel: string;
|
|
62
|
+
}
|
|
63
|
+
>
|
|
64
|
+
>;
|
|
65
|
+
}
|
|
66
|
+
|
|
67
|
+
/** Dimensions carried across from the kit's declaration. */
|
|
68
|
+
export const KIT_MAPPED_DIMENSIONS: readonly string[] = [
|
|
69
|
+
"gateId",
|
|
70
|
+
"acceptedEnvelopeSpecs",
|
|
71
|
+
"identityVersions",
|
|
72
|
+
// Derived from the declaration: the kit names exactly one identity version,
|
|
73
|
+
// and a recorder must accept the identity it computes.
|
|
74
|
+
"computingIdentityVersion",
|
|
75
|
+
"providers.ids",
|
|
76
|
+
"obligations.ids",
|
|
77
|
+
"obligations.providers",
|
|
78
|
+
"obligations.acceptedPayloadSpecs",
|
|
79
|
+
"obligations.minimumAttestationLevel",
|
|
80
|
+
];
|
|
81
|
+
|
|
82
|
+
/**
|
|
83
|
+
* Dimensions the kit does not declare, each with the reason its default is safe.
|
|
84
|
+
* `invariant-constrained` means the value is not a free choice; every other
|
|
85
|
+
* entry is a dimension no vector expectation can turn on.
|
|
86
|
+
*/
|
|
87
|
+
export const KIT_DEFAULTED_DIMENSIONS: Readonly<Record<string, "vector-independent" | "invariant-constrained">> = {
|
|
88
|
+
baseRef: "vector-independent",
|
|
89
|
+
storageNamespace: "vector-independent",
|
|
90
|
+
reviewNeutral: "invariant-constrained",
|
|
91
|
+
recordNeutral: "vector-independent",
|
|
92
|
+
pathClassification: "vector-independent",
|
|
93
|
+
sensitivePaths: "vector-independent",
|
|
94
|
+
activationThreshold: "vector-independent",
|
|
95
|
+
agentEnvSignals: "vector-independent",
|
|
96
|
+
ciPolicies: "vector-independent",
|
|
97
|
+
ciPolicyEnvKey: "vector-independent",
|
|
98
|
+
preparationWiringPaths: "vector-independent",
|
|
99
|
+
deliveryRecordPath: "vector-independent",
|
|
100
|
+
deliveryRecordVerification: "vector-independent",
|
|
101
|
+
"providers.findingCodes": "vector-independent",
|
|
102
|
+
"obligations.activation.kind": "vector-independent",
|
|
103
|
+
"obligations.activation.sensitiveGroupIds": "vector-independent",
|
|
104
|
+
"obligations.activation.relevantBinaryChangeActivates": "vector-independent",
|
|
105
|
+
"obligations.activation.relevantZeroLineChangeActivates": "vector-independent",
|
|
106
|
+
"obligations.freshness": "vector-independent",
|
|
107
|
+
"obligations.providerPolicy": "vector-independent",
|
|
108
|
+
"obligations.allowedResolutionKinds": "vector-independent",
|
|
109
|
+
"obligations.humanWaiverAllowed": "vector-independent",
|
|
110
|
+
"obligations.ciDelegationPolicyIds": "vector-independent",
|
|
111
|
+
"obligations.remediation": "vector-independent",
|
|
112
|
+
"obligations.waivableCodes": "vector-independent",
|
|
113
|
+
"obligations.nonWaivableCodes": "vector-independent",
|
|
114
|
+
};
|
|
115
|
+
|
|
116
|
+
/** Tokens whose absence four reject vectors depend on. */
|
|
117
|
+
export const KIT_NEGATIVE_SPACE: readonly string[] = [
|
|
118
|
+
"qa.exercised",
|
|
119
|
+
"review.green/0",
|
|
120
|
+
"unknown.provider",
|
|
121
|
+
"deliverable-tree/v0",
|
|
122
|
+
];
|
|
123
|
+
|
|
124
|
+
/**
|
|
125
|
+
* The gate's structural finding codes, restated so the adapter's classification
|
|
126
|
+
* is a decision this file makes rather than a set it inherits. A structural code
|
|
127
|
+
* added to the contract shows up here as an unclassified-code finding at load
|
|
128
|
+
* time, which is the intended way to learn about it.
|
|
129
|
+
*/
|
|
130
|
+
const STRUCTURAL_CODES: readonly string[] = [
|
|
131
|
+
"review_evidence_missing",
|
|
132
|
+
"stale_evidence",
|
|
133
|
+
"evidence_not_green",
|
|
134
|
+
"unresolved_actionable_findings",
|
|
135
|
+
"ambiguous_records",
|
|
136
|
+
"malformed_record",
|
|
137
|
+
"unknown_provider",
|
|
138
|
+
"live_provider_missing",
|
|
139
|
+
"ambiguous_live_provider",
|
|
140
|
+
"live_provider_failed",
|
|
141
|
+
"resolution_not_allowed",
|
|
142
|
+
];
|
|
143
|
+
|
|
144
|
+
function requireRecord(value: unknown, label: string): Record<string, unknown> {
|
|
145
|
+
if (typeof value !== "object" || value === null || Array.isArray(value)) {
|
|
146
|
+
throw new Error(`${label} must be an object.`);
|
|
147
|
+
}
|
|
148
|
+
return value as Record<string, unknown>;
|
|
149
|
+
}
|
|
150
|
+
|
|
151
|
+
/**
|
|
152
|
+
* Maps a parsed kit repo-config into a harness config. Pure: the caller supplies
|
|
153
|
+
* the parsed value, so the mapping is testable against a synthetic declaration
|
|
154
|
+
* without touching the vendored kit.
|
|
155
|
+
*/
|
|
156
|
+
export function adaptKitRepoConfig(raw: unknown): HarnessConfig {
|
|
157
|
+
const declared = requireRecord(raw, "kit repo-config");
|
|
158
|
+
const obligations = requireRecord(declared["obligations"], "kit repo-config obligations");
|
|
159
|
+
|
|
160
|
+
const identityVersions = declared["identityVersions"] as readonly string[];
|
|
161
|
+
const providerIds = [
|
|
162
|
+
...new Set(
|
|
163
|
+
Object.values(obligations).flatMap((obligation) => (requireRecord(obligation, "kit obligation")["providers"] as readonly string[]) ?? []),
|
|
164
|
+
),
|
|
165
|
+
];
|
|
166
|
+
|
|
167
|
+
const input: HarnessConfigInput = {
|
|
168
|
+
// ── Mapped ──
|
|
169
|
+
gateId: declared["gateId"] as string,
|
|
170
|
+
acceptedEnvelopeSpecs: declared["acceptedEnvelopeSpecs"] as readonly string[],
|
|
171
|
+
identityVersions,
|
|
172
|
+
// The kit declares one identity version; a recorder must accept the identity
|
|
173
|
+
// it computes, so the sole declared version is the computing one.
|
|
174
|
+
computingIdentityVersion: identityVersions[0] as string,
|
|
175
|
+
// The kit registers providers per obligation and declares no finding codes
|
|
176
|
+
// for them, so the registry is their union with an empty code surface. That
|
|
177
|
+
// leaves the emittable universe equal to the structural registry, which is
|
|
178
|
+
// what the classification below partitions.
|
|
179
|
+
providers: providerIds.map((id) => ({ id, findingCodes: [] })),
|
|
180
|
+
obligations: Object.entries(obligations).map(([id, declaredObligation]) => {
|
|
181
|
+
const obligation = requireRecord(declaredObligation, `kit obligation ${id}`);
|
|
182
|
+
return {
|
|
183
|
+
id,
|
|
184
|
+
acceptedPayloadSpecs: obligation["acceptedPayloadSpecs"] as readonly string[],
|
|
185
|
+
providers: obligation["providers"] as readonly string[],
|
|
186
|
+
minimumAttestationLevel: obligation["minimumAttestationLevel"] as "self",
|
|
187
|
+
// ── Defaulted, per obligation ──
|
|
188
|
+
activation: { kind: "relevant_change" },
|
|
189
|
+
freshness: "exact_candidate",
|
|
190
|
+
allowedResolutionKinds: ["satisfied_evidence", "not_applicable"],
|
|
191
|
+
humanWaiverAllowed: false,
|
|
192
|
+
ciDelegationPolicyIds: [],
|
|
193
|
+
remediation: {
|
|
194
|
+
default: [
|
|
195
|
+
{
|
|
196
|
+
id: "submit-evidence",
|
|
197
|
+
kind: "manual_action",
|
|
198
|
+
summary: `Submit a manifest claiming ${id} for the current candidate.`,
|
|
199
|
+
},
|
|
200
|
+
],
|
|
201
|
+
},
|
|
202
|
+
waivableCodes: [],
|
|
203
|
+
nonWaivableCodes: STRUCTURAL_CODES,
|
|
204
|
+
};
|
|
205
|
+
}),
|
|
206
|
+
// ── Defaulted ──
|
|
207
|
+
baseRef: "origin/main",
|
|
208
|
+
storageNamespace: "delivery-harness/",
|
|
209
|
+
reviewNeutral: [...DELIVERABLE_TREE_V1_NARRATION_SET],
|
|
210
|
+
recordNeutral: [{ prefix: "docs/reports/" }],
|
|
211
|
+
pathClassification: {
|
|
212
|
+
generated: [{ kind: "prefix", value: "generated/" }],
|
|
213
|
+
test: [{ kind: "glob", value: "**/*.test.ts" }],
|
|
214
|
+
lockfile: [{ kind: "glob", value: "**/package-lock.json" }],
|
|
215
|
+
},
|
|
216
|
+
sensitivePaths: [],
|
|
217
|
+
activationThreshold: 1,
|
|
218
|
+
agentEnvSignals: [],
|
|
219
|
+
ciPolicies: [],
|
|
220
|
+
ciPolicyEnvKey: "DELIVERY_HARNESS_CI_POLICY",
|
|
221
|
+
preparationWiringPaths: ["harness.config.ts"],
|
|
222
|
+
deliveryRecordPath: "docs/reports/delivery-record.json",
|
|
223
|
+
deliveryRecordVerification: { baseMovement: "stale" },
|
|
224
|
+
};
|
|
225
|
+
|
|
226
|
+
return defineHarnessConfig(input);
|
|
227
|
+
}
|
|
228
|
+
|
|
229
|
+
/** Repo-relative path of the vendored declaration this adapter reads. */
|
|
230
|
+
export const KIT_REPO_CONFIG_PATH = "packages/conformance/vectors/context/repo-config.json";
|
|
231
|
+
|
|
232
|
+
export function readKitRepoConfig(): KitRepoConfig {
|
|
233
|
+
const here = path.dirname(fileURLToPath(import.meta.url));
|
|
234
|
+
const file = path.join(here, "..", "vectors", "context", "repo-config.json");
|
|
235
|
+
return JSON.parse(readFileSync(file, "utf8")) as KitRepoConfig;
|
|
236
|
+
}
|
|
237
|
+
|
|
238
|
+
/** The kit's own configuration, loaded and validated. */
|
|
239
|
+
export function loadKitRepoConfig(): HarnessConfig {
|
|
240
|
+
return adaptKitRepoConfig(readKitRepoConfig());
|
|
241
|
+
}
|
|
242
|
+
|
|
243
|
+
/**
|
|
244
|
+
* Which negative-space tokens a config has let in. Empty is the only correct
|
|
245
|
+
* answer for a config the kit is evaluated under.
|
|
246
|
+
*/
|
|
247
|
+
export function negativeSpaceViolations(config: HarnessConfig): readonly string[] {
|
|
248
|
+
const declared = JSON.stringify(config);
|
|
249
|
+
return KIT_NEGATIVE_SPACE.filter((token) => declared.includes(token));
|
|
250
|
+
}
|
package/package.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
1
|
+
{
|
|
2
|
+
"name": "@agent-delivery-harness/conformance",
|
|
3
|
+
"version": "0.1.0",
|
|
4
|
+
"type": "module",
|
|
5
|
+
"description": "delivery-evidence/1 conformance kit (89 golden vectors) and its generator",
|
|
6
|
+
"license": "Apache-2.0",
|
|
7
|
+
"engines": {
|
|
8
|
+
"node": ">=22"
|
|
9
|
+
},
|
|
10
|
+
"dependencies": {
|
|
11
|
+
"@agent-delivery-harness/kernel": "0.1.0"
|
|
12
|
+
},
|
|
13
|
+
"exports": {
|
|
14
|
+
".": "./src/index.ts"
|
|
15
|
+
},
|
|
16
|
+
"files": [
|
|
17
|
+
"src",
|
|
18
|
+
"!src/**/*.test.ts",
|
|
19
|
+
"vectors",
|
|
20
|
+
"fixtures/repo-config-adapter.ts",
|
|
21
|
+
"LICENSE",
|
|
22
|
+
"NOTICE"
|
|
23
|
+
]
|
|
24
|
+
}
|