@hap-labs/human-agent-paradigm 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/CHANGELOG.md +21 -0
- package/LICENSE +21 -0
- package/README.md +60 -0
- package/README.zh-CN.md +52 -0
- package/assets/EN/CONFORMANCE_CHECKLIST.md +211 -0
- package/assets/EN/CONTRACT_TEMPLATE.md +241 -0
- package/assets/EN/DECISION_REQUEST_TEMPLATE.md +157 -0
- package/assets/EN/DELIVERY_REPORT_TEMPLATE.md +188 -0
- package/assets/EN/DERIVED_SPECIFICATION.md +607 -0
- package/assets/EN/HUMAN_AGENT_PARADIGM.md +390 -0
- package/assets/ZH_CN/CONFORMANCE_CHECKLIST.md +192 -0
- package/assets/ZH_CN/CONTRACT_TEMPLATE.md +216 -0
- package/assets/ZH_CN/DECISION_REQUEST_TEMPLATE.md +140 -0
- package/assets/ZH_CN/DELIVERY_REPORT_TEMPLATE.md +199 -0
- package/assets/ZH_CN/DERIVED_SPECIFICATION.md +497 -0
- package/assets/ZH_CN/HUMAN_AGENT_PARADIGM.md +285 -0
- package/package.json +38 -0
- package/scripts/python/repo_governance_check/__init__.py +3 -0
- package/scripts/python/repo_governance_check/__main__.py +6 -0
- package/scripts/python/repo_governance_check/checks.py +359 -0
- package/scripts/python/repo_governance_check/cli.py +115 -0
- package/scripts/python/repo_governance_check/config.py +33 -0
- package/scripts/python/repo_governance_check/governance_check_config.json +166 -0
- package/scripts/python/repo_governance_check/report.py +79 -0
|
@@ -0,0 +1,241 @@
|
|
|
1
|
+
# Contract Template (execution layer)
|
|
2
|
+
|
|
3
|
+
> English translation of the frozen document `CONTRACT_TEMPLATE.md`. The Simplified Chinese original is authoritative
|
|
4
|
+
> for governance and is the signing copy; this English copy is provided for agent reading, understanding, and reference.
|
|
5
|
+
> Version markers mirror the original (v1.0). Upstream references point to `HUMAN_AGENT_PARADIGM.md` and
|
|
6
|
+
> `DERIVED_SPECIFICATION.md`.
|
|
7
|
+
|
|
8
|
+
Version: v1.0
|
|
9
|
+
Hierarchy: constitution `HUMAN_AGENT_PARADIGM.md` v1.0 → derived specification `DERIVED_SPECIFICATION.md` v1.0 → this
|
|
10
|
+
template
|
|
11
|
+
Status: execution-layer template · produced with the Owner's authorization
|
|
12
|
+
Applicable scope: the P3 contract-establishment phase; the first basis of judgment for all subsequent work (provided it
|
|
13
|
+
does not conflict with the constitution or the derived specification)
|
|
14
|
+
|
|
15
|
+
## Usage Rules
|
|
16
|
+
|
|
17
|
+
1. This template is filled in by the Agent at P3 and confirmed and signed by the human (the Owner); design and
|
|
18
|
+
implementation must not start before the contract is established.
|
|
19
|
+
2. None of C1–C10 may be omitted; small tasks may shorten sections, but every section must answer "is it defined, what
|
|
20
|
+
is the criterion, what is the evidence".
|
|
21
|
+
3. All thresholds, criteria, and budgets are proposed by the Agent and confirmed by the human; the Agent may not
|
|
22
|
+
unilaterally relax them, and later changes must go through C10.
|
|
23
|
+
4. The contract ID is unique and versions are traceable; every change must record the reason, affected clauses, and the
|
|
24
|
+
human's approval.
|
|
25
|
+
5. This contract must not conflict with the constitution or the derived specification; on conflict, the higher-level
|
|
26
|
+
norms prevail and the matter returns to the contract process.
|
|
27
|
+
|
|
28
|
+
---
|
|
29
|
+
|
|
30
|
+
## Contract Metadata
|
|
31
|
+
|
|
32
|
+
| Field | Content |
|
|
33
|
+
|:----------------------------|:----------------------------------------------------|
|
|
34
|
+
| Contract ID | |
|
|
35
|
+
| Version | |
|
|
36
|
+
| Product/task name | |
|
|
37
|
+
| Upstream intent record | |
|
|
38
|
+
| Direction-decision record | |
|
|
39
|
+
| Constitutional traceability | constitution 5.2 P3, 3.2, 7.3; derived spec 2.1–2.4 |
|
|
40
|
+
| Status | `Draft` / `Signed` / `Amended` / `Terminated` |
|
|
41
|
+
|
|
42
|
+
---
|
|
43
|
+
|
|
44
|
+
## C1 Background and Intent
|
|
45
|
+
|
|
46
|
+
**Original intent (stated by the human):**
|
|
47
|
+
|
|
48
|
+
**Agent restatement (requires human confirmation):**
|
|
49
|
+
|
|
50
|
+
**"What matters more" ordering:** 1. ______ 2. ______ 3. ______
|
|
51
|
+
|
|
52
|
+
**Key preferences and values:**
|
|
53
|
+
|
|
54
|
+
**Differences awaiting confirmation (if any):**
|
|
55
|
+
|
|
56
|
+
- [ ] The human has confirmed the restatement and ordering above.
|
|
57
|
+
|
|
58
|
+
---
|
|
59
|
+
|
|
60
|
+
## C2 Scope and Non-Goals
|
|
61
|
+
|
|
62
|
+
| No. | In-scope deliverable | Judgeable completion criterion (cite C4/C5 criterion IDs) |
|
|
63
|
+
|:----|:---------------------|:----------------------------------------------------------|
|
|
64
|
+
| S1 | | |
|
|
65
|
+
| S2 | | |
|
|
66
|
+
|
|
67
|
+
**Explicit non-goals (not delivered this run):**
|
|
68
|
+
|
|
69
|
+
**Product form:** (what the deliverable, runnable, usable outcome is; not limited to software)
|
|
70
|
+
|
|
71
|
+
---
|
|
72
|
+
|
|
73
|
+
## C3 Constraints and Bottom Lines
|
|
74
|
+
|
|
75
|
+
| Category | Constraint | Non-crossable criterion | Handling on conflict |
|
|
76
|
+
|:------------------------|:-----------|:------------------------|:---------------------|
|
|
77
|
+
| Time | | | |
|
|
78
|
+
| Cost | | | |
|
|
79
|
+
| Risk | | | |
|
|
80
|
+
| Safety | | | |
|
|
81
|
+
| Compliance | | | |
|
|
82
|
+
| Privacy/confidentiality | | | |
|
|
83
|
+
| Technology/resources | | | |
|
|
84
|
+
|
|
85
|
+
**Bottom-line statement:** safety and compliance outrank all efficiency and need metrics; human authorization never
|
|
86
|
+
overrides law or basic safety bottom lines.
|
|
87
|
+
|
|
88
|
+
---
|
|
89
|
+
|
|
90
|
+
## C4 Reliability Criteria
|
|
91
|
+
|
|
92
|
+
**Definition of done (DoD):** a "completed" claim is allowed only when all of the following hold:
|
|
93
|
+
|
|
94
|
+
1. the corresponding C2 deliverable runs/is usable;
|
|
95
|
+
2. every completion claim carries real evidence (evidence IDs pointing into the evidence package);
|
|
96
|
+
3. the C6 multi-view review passed with no unresolved in-contract defects;
|
|
97
|
+
4. failures and incomplete items are listed truthfully.
|
|
98
|
+
|
|
99
|
+
**Claim–evidence mapping (concretized for this project):**
|
|
100
|
+
|
|
101
|
+
| Claim type | Required evidence form | Evidence destination |
|
|
102
|
+
|:-----------------------------|:-------------------------------------------------------------------------|:---------------------|
|
|
103
|
+
| Feature/deliverable complete | runnable/usable artifact + run/use/test records | evidence package |
|
|
104
|
+
| Quality met | review/test results + defect records + fix/re-verification records | evidence package |
|
|
105
|
+
| State report | snapshot of current real state + differences from previous state | evidence package |
|
|
106
|
+
| Performance met | measurement plan under contract conditions + results + environment notes | evidence package |
|
|
107
|
+
| Asset reused | source + verification records + post-reuse results | evidence package |
|
|
108
|
+
|
|
109
|
+
**State-sync method:** (when to sync, what to sync, in which objective metrics)
|
|
110
|
+
|
|
111
|
+
**Failure-reporting method:** (how failures/unmet standards are reported once found, content and time limits; vague
|
|
112
|
+
wording forbidden)
|
|
113
|
+
|
|
114
|
+
**Prohibited behaviors:** reporting things that did not happen or were not completed; substituting simulation,
|
|
115
|
+
description, or imagination for real output; fabricating evidence or citing nonexistent things; replacing checkable
|
|
116
|
+
evidence with self-assessment.
|
|
117
|
+
|
|
118
|
+
---
|
|
119
|
+
|
|
120
|
+
## C5 Excellence Criteria (five dimensions, each judgeable)
|
|
121
|
+
|
|
122
|
+
**Filling rule:** every criterion must be writable as "object + condition + observable result + threshold/criterion +
|
|
123
|
+
evidence source"; adjectives alone are not allowed.
|
|
124
|
+
|
|
125
|
+
| Criterion ID | Dimension | Object and condition | Observable result | Threshold/criterion | Evidence source and method |
|
|
126
|
+
|:-------------|:-----------------------|:---------------------|:------------------|:--------------------|:---------------------------|
|
|
127
|
+
| E6.1.x | Sound design | | | | |
|
|
128
|
+
| E6.2.x | Complete functionality | | | | |
|
|
129
|
+
| E6.3.x | Elegant implementation | | | | |
|
|
130
|
+
| E6.4.x | Perfect experience | | | | |
|
|
131
|
+
| E6.5.x | Excellent performance | | | | |
|
|
132
|
+
|
|
133
|
+
- [ ] All five dimension criteria are defined; no "adjective as acceptance" blank items.
|
|
134
|
+
- [ ] The human confirmed all thresholds; the Agent did not unilaterally relax them.
|
|
135
|
+
|
|
136
|
+
---
|
|
137
|
+
|
|
138
|
+
## C6 Multi-View Review Requirements
|
|
139
|
+
|
|
140
|
+
| Perspective | Problems that must be sought | Review-record requirements |
|
|
141
|
+
|:------------|:-------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------|
|
|
142
|
+
| Product | intent drift, scope omissions, non-goal bloat, value proposition not holding | independent standpoint, defect-finding method, findings, conclusion |
|
|
143
|
+
| User | whether target users complete independently, paths, feedback, error recovery, friction | same |
|
|
144
|
+
| Engineering | design soundness, implementation craft, maintainability, quality debt, boundaries | same |
|
|
145
|
+
| Adversarial | assume completion claims false; find counterexamples, exceptional paths, safety/compliance holes | same |
|
|
146
|
+
|
|
147
|
+
**Independence:** review standpoint, method, and records are independent of the production process of the content under
|
|
148
|
+
review; the same entity executing a review must explicitly switch standpoints and use independent methods.
|
|
149
|
+
|
|
150
|
+
**Defect closure:** Blockers/Majors must be fixed and re-verified; Minors should be fixed; any defect chosen not to fix
|
|
151
|
+
must reopen the contract, be human-approved out of scope, and be explicitly disclosed.
|
|
152
|
+
|
|
153
|
+
**Exit criteria:** all C5 criteria met and no known unresolved in-contract defect.
|
|
154
|
+
|
|
155
|
+
---
|
|
156
|
+
|
|
157
|
+
## C7 Autonomy Budget
|
|
158
|
+
|
|
159
|
+
| Item | Content |
|
|
160
|
+
|:-----------------------|:--------------------|
|
|
161
|
+
| Authorized scope | |
|
|
162
|
+
| Risk threshold | escalate above this |
|
|
163
|
+
| Cost cap | escalate above this |
|
|
164
|
+
| Self-decidable matters | |
|
|
165
|
+
| Must-escalate matters | |
|
|
166
|
+
| Boundary-touch action | |
|
|
167
|
+
|
|
168
|
+
**Rules:** no asking inside the boundaries; escalate immediately on touching a boundary; the autonomy budget may only be
|
|
169
|
+
adjusted by the human in a contract or explicit authorization; in-budget self-decisions must not conflict with the
|
|
170
|
+
contract or effectively lower the reliability or excellence standard.
|
|
171
|
+
|
|
172
|
+
---
|
|
173
|
+
|
|
174
|
+
## C8 Deliverables and Evidence List
|
|
175
|
+
|
|
176
|
+
| Category | Concrete deliverable | Corresponding evidence/location |
|
|
177
|
+
|:-----------------|:---------------------|:--------------------------------|
|
|
178
|
+
| Product itself | | |
|
|
179
|
+
| Evidence package | | |
|
|
180
|
+
| Reusable assets | | |
|
|
181
|
+
| Delivery report | | |
|
|
182
|
+
|
|
183
|
+
---
|
|
184
|
+
|
|
185
|
+
## C9 Acceptance Method
|
|
186
|
+
|
|
187
|
+
**Sole acceptance basis:** this contract (C2–C5, C6, C8).
|
|
188
|
+
|
|
189
|
+
**Acceptance actions:** the human verifies evidence item by item against the contract and signs the acceptance
|
|
190
|
+
conclusion and time; the Agent may not substitute self-justification, self-assessment, or "looks compliant" for contract
|
|
191
|
+
requirements.
|
|
192
|
+
|
|
193
|
+
**Forbidden:** adding out-of-contract requirements ad hoc at acceptance; new requirements must return to P0/P3 to be
|
|
194
|
+
re-aligned and re-priced.
|
|
195
|
+
|
|
196
|
+
---
|
|
197
|
+
|
|
198
|
+
## C10 Change, Termination, and Failure Handling
|
|
199
|
+
|
|
200
|
+
| Trigger | Handling |
|
|
201
|
+
|:-----------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------|
|
|
202
|
+
| Human changes intent/constraints | return to P0/P3; re-price affected parts |
|
|
203
|
+
| Facts conflict with the contract | stop, report truthfully, submit a redirect/reopen proposal |
|
|
204
|
+
| Contract does not cover a direction-affecting matter | return to P3 to amend the contract |
|
|
205
|
+
| Deviation from a confirmed direction | must be proposed and approved first; acting first and telling later — or acting and staying silent — is forbidden |
|
|
206
|
+
| Reliability and excellence cannot both be met within constraints | stop, escalate the trade-off, ask for constraint adjustment; silent trade-offs forbidden |
|
|
207
|
+
| Human stops the project | deliver what was actually completed, evidence, and an incomplete-state statement; never claim incomplete items as complete |
|
|
208
|
+
| Acceptance fails | fix per the contract defect list, or reopen the contract to adjust criteria |
|
|
209
|
+
|
|
210
|
+
---
|
|
211
|
+
|
|
212
|
+
## Traceability and Effectiveness
|
|
213
|
+
|
|
214
|
+
| Contract section | Constitutional clauses served | Derived-spec clauses served |
|
|
215
|
+
|:-----------------|:------------------------------|:----------------------------|
|
|
216
|
+
| C1 | 1.2 S3, 4.1, 5.2 P0 | 3.1 |
|
|
217
|
+
| C2 | 1.4, 5.2 P3 | 2.1 |
|
|
218
|
+
| C3 | 3.4, 7.3, 7.4 | 2.1, 7.3 |
|
|
219
|
+
| C4 | 3.1, 5.5 | 2.1, 5 |
|
|
220
|
+
| C5 | 3.2, 6.1–6.6 | 2.2, 2.3, 6 |
|
|
221
|
+
| C6 | 3.2, 6.6 | 6.2–6.5 |
|
|
222
|
+
| C7 | 7.3, A4 | 7.3 |
|
|
223
|
+
| C8 | 1.3, 5.5 | 1.1 |
|
|
224
|
+
| C9 | 5.5 | 3.8 |
|
|
225
|
+
| C10 | 5.3, 5.4, 1.4 | 2.4, 3.11 |
|
|
226
|
+
|
|
227
|
+
**Pre-delivery gate:** execute《Conformance Self-Check Checklist》v1.0 groups A–I with all ★ items passing; when claiming
|
|
228
|
+
"the paradigm has been implemented", J1–J7 must also all pass.
|
|
229
|
+
|
|
230
|
+
---
|
|
231
|
+
|
|
232
|
+
## Signature and Revision
|
|
233
|
+
|
|
234
|
+
| Role | Content | Signature | Time |
|
|
235
|
+
|:---------------------------|:------------------------------------------------------------------------------|:----------|:-----|
|
|
236
|
+
| Agent drafter | disclose pros, cons, costs, and standard strength completely and without bias | | |
|
|
237
|
+
| Human (Owner) confirmation | confirm intent, constraints, criteria, and budget | | |
|
|
238
|
+
|
|
239
|
+
| Version | Date | Change | Affected clauses | Human approval |
|
|
240
|
+
|:--------|:-----|:-------|:-----------------|:---------------|
|
|
241
|
+
| | | | | |
|
|
@@ -0,0 +1,157 @@
|
|
|
1
|
+
# Decision Request and Necessary Information Request Template (execution layer)
|
|
2
|
+
|
|
3
|
+
> English translation of the frozen document `DECISION_REQUEST_TEMPLATE.md`. The Simplified Chinese original is
|
|
4
|
+
> authoritative for governance; this English copy is provided for agent reading, understanding, and reference. Version
|
|
5
|
+
> markers mirror the original (v1.0). Upstream references point to `HUMAN_AGENT_PARADIGM.md` and
|
|
6
|
+
> `DERIVED_SPECIFICATION.md`.
|
|
7
|
+
|
|
8
|
+
Version: v1.0
|
|
9
|
+
Hierarchy: constitution `HUMAN_AGENT_PARADIGM.md` v1.0 → derived specification `DERIVED_SPECIFICATION.md` v1.0 → this
|
|
10
|
+
template
|
|
11
|
+
Status: execution-layer template · produced with the Owner's authorization
|
|
12
|
+
Applicable scope: any interaction during P0–P7 that needs a human response; Part A is for necessary decisions
|
|
13
|
+
(preference/value, authorization/responsibility); Part B is for necessary information completion in P0/P1
|
|
14
|
+
|
|
15
|
+
## Usage Rules
|
|
16
|
+
|
|
17
|
+
1. Run the necessity test first: anything the Agent could ascertain, verify, or cover under existing authorization must
|
|
18
|
+
not be sent to the human.
|
|
19
|
+
2. "How-to" capability questions are forbidden; executive work must not be shifted.
|
|
20
|
+
3. Part A decision requests must carry all six elements (background summary, substantive options, recommendation,
|
|
21
|
+
reason, default choice, decision impact) and state a response window.
|
|
22
|
+
4. Preference/value types: when the human is late in responding, the default may proceed with a disclosure trace;
|
|
23
|
+
authorization/responsibility types: no default — wait for the explicit decision.
|
|
24
|
+
5. Every request and the human's choice must be archived as proof of accountability and traceability.
|
|
25
|
+
|
|
26
|
+
---
|
|
27
|
+
|
|
28
|
+
## Part A Decision Request
|
|
29
|
+
|
|
30
|
+
### A0 Metadata
|
|
31
|
+
|
|
32
|
+
| Field | Content |
|
|
33
|
+
|:--------------|:------------------------------------------------------------|
|
|
34
|
+
| Request ID | |
|
|
35
|
+
| Contract ID | |
|
|
36
|
+
| Current phase | |
|
|
37
|
+
| Request type | `P preference/value` / `A authorization/responsibility` |
|
|
38
|
+
| Risk level | `High/Medium/Low` (drives the window and escalation method) |
|
|
39
|
+
| Created at | |
|
|
40
|
+
|
|
41
|
+
### A1 Necessity Test (mandatory; do not send if it fails)
|
|
42
|
+
|
|
43
|
+
1. Can the matter be resolved by the Agent through information gathering, reasoning, verification, or existing
|
|
44
|
+
authorization?
|
|
45
|
+
- `Yes` → **stop sending; the Agent decides on its own.**
|
|
46
|
+
- `No` → continue to step 2.
|
|
47
|
+
2. Is the matter essentially the human's preference/value, or authorization/responsibility? (choose exactly one and
|
|
48
|
+
state the basis)
|
|
49
|
+
3. If borderline: wrong self-decision costly and irreversible → escalate; cheap and reversible → decide and disclose
|
|
50
|
+
afterwards.
|
|
51
|
+
4. Why this request cannot be self-decided by the Agent:
|
|
52
|
+
|
|
53
|
+
### A2 Background Summary
|
|
54
|
+
|
|
55
|
+
**Enough for an independent third party to understand the facts needed for the decision, without irrelevant
|
|
56
|
+
information:**
|
|
57
|
+
|
|
58
|
+
**Contract clauses directly related to this decision:**
|
|
59
|
+
|
|
60
|
+
### A3 Substantive Options
|
|
61
|
+
|
|
62
|
+
**At least two for type P; for type A at least "approve/not approve". Each option states key differences, benefits,
|
|
63
|
+
cost, and risk.**
|
|
64
|
+
|
|
65
|
+
| Option | Content | Key difference | Benefit | Cost/risk | Impact on standard strength |
|
|
66
|
+
|:---------|:--------|:---------------|:--------|:----------|:----------------------------|
|
|
67
|
+
| Option 1 | | | | | |
|
|
68
|
+
| Option 2 | | | | | |
|
|
69
|
+
| (if any) | | | | | |
|
|
70
|
+
|
|
71
|
+
**Substantive-difference check:** the options differ discernibly on key dimensions such as scope/path/quality
|
|
72
|
+
strategy/cost-risk/experience orientation — not rewording.
|
|
73
|
+
|
|
74
|
+
### A4 Recommendation and Reasons
|
|
75
|
+
|
|
76
|
+
| Field | Content |
|
|
77
|
+
|:---------------------------------------|:--------|
|
|
78
|
+
| Recommended option | |
|
|
79
|
+
| Recommendation reasons | |
|
|
80
|
+
| Cost of the recommendation | |
|
|
81
|
+
| Impact of rejecting the recommendation | |
|
|
82
|
+
|
|
83
|
+
### A5 Default Choice and Timeout Handling
|
|
84
|
+
|
|
85
|
+
| Field | Content |
|
|
86
|
+
|:----------------------------------|:--------------------------------------------------------------------------------------------|
|
|
87
|
+
| Default option | type P may fill one; type A must fill "no default — wait for the explicit decision" |
|
|
88
|
+
| Default activation condition | only for preference/value types when the human has not responded within the response window |
|
|
89
|
+
| Disclosure after default proceeds | leave a trace in the decision records and the delivery report |
|
|
90
|
+
|
|
91
|
+
### A6 Decision Impact
|
|
92
|
+
|
|
93
|
+
| Option | Scope impact | Cost impact | Quality impact | Risk impact | Time impact |
|
|
94
|
+
|:---------|:-------------|:------------|:---------------|:------------|:------------|
|
|
95
|
+
| Option 1 | | | | | |
|
|
96
|
+
| Option 2 | | | | | |
|
|
97
|
+
|
|
98
|
+
### A7 Response Window
|
|
99
|
+
|
|
100
|
+
| Field | Content |
|
|
101
|
+
|:--------------------------|:------------------------------------------------------------------------------------|
|
|
102
|
+
| Suggested response window | |
|
|
103
|
+
| Window basis | risk, cost, time sensitivity |
|
|
104
|
+
| Timeout handling | type P: proceed on the A5 default with a trace; type A: keep waiting, never default |
|
|
105
|
+
|
|
106
|
+
### A8 The Human's Decision and Trace
|
|
107
|
+
|
|
108
|
+
| Field | Content |
|
|
109
|
+
|:-----------------------------|:-------------------------------------------------------------|
|
|
110
|
+
| Human's choice | |
|
|
111
|
+
| Basis of the choice | |
|
|
112
|
+
| Decision-impact confirmation | |
|
|
113
|
+
| Decision time | |
|
|
114
|
+
| Traceability references | constitution 4.3/4.4; derived spec 4.1–4.4; contract clauses |
|
|
115
|
+
|
|
116
|
+
---
|
|
117
|
+
|
|
118
|
+
## Part B Necessary Information Request (P0/P1 only)
|
|
119
|
+
|
|
120
|
+
### B0 Pre-checks (do not send if they fail)
|
|
121
|
+
|
|
122
|
+
- [ ] This request happens in P0/P1.
|
|
123
|
+
- [ ] Self-service retrieval was attempted and at least one tried channel is recorded.
|
|
124
|
+
- [ ] The information is necessary to proceed and cannot be replaced by reasoning, verification, or existing
|
|
125
|
+
authorization.
|
|
126
|
+
- [ ] Contains no "how-to" and requires no executive work from the human.
|
|
127
|
+
|
|
128
|
+
### B1 Request Content
|
|
129
|
+
|
|
130
|
+
| Field | Content |
|
|
131
|
+
|:---------------------------------------------------------------|:--------|
|
|
132
|
+
| Request ID | |
|
|
133
|
+
| Contract/intent record | |
|
|
134
|
+
| Exact missing information | |
|
|
135
|
+
| Which factual judgment/contract clause this information serves | |
|
|
136
|
+
| Self-service channels already tried | |
|
|
137
|
+
| Why it cannot be obtained independently | |
|
|
138
|
+
| How the Agent will degrade if it cannot be provided | |
|
|
139
|
+
|
|
140
|
+
### B2 The Human's Answer and Trace
|
|
141
|
+
|
|
142
|
+
| Field | Content |
|
|
143
|
+
|:----------------------------------|:------------------------------------------------|
|
|
144
|
+
| Information provided by the human | |
|
|
145
|
+
| Record the information enters | intent record / fact-and-constraint baseline |
|
|
146
|
+
| Answer time | |
|
|
147
|
+
| Traceability references | constitution 4.2, 5.2 P1; derived spec 4.1, 4.4 |
|
|
148
|
+
|
|
149
|
+
---
|
|
150
|
+
|
|
151
|
+
## Pre-Send Self-Check
|
|
152
|
+
|
|
153
|
+
- [ ] Not a "how-to" question.
|
|
154
|
+
- [ ] Not something the Agent could self-decide.
|
|
155
|
+
- [ ] Part A has all six elements; type P has ≥2 substantively different options; type A has no default.
|
|
156
|
+
- [ ] Response window and timeout handling stated.
|
|
157
|
+
- [ ] Disclosure complete and unbiased; no information advantage used to induce weaker standards.
|
|
@@ -0,0 +1,188 @@
|
|
|
1
|
+
# Delivery Report Template (execution layer)
|
|
2
|
+
|
|
3
|
+
> English translation of the frozen document `DELIVERY_REPORT_TEMPLATE.md`. The Simplified Chinese original is
|
|
4
|
+
> authoritative for governance and is the signing copy; this English copy is provided for agent reading, understanding,
|
|
5
|
+
> and reference. Version markers mirror the original (v1.0). The checklist reference points to
|
|
6
|
+
> `CONFORMANCE_CHECKLIST.md`
|
|
7
|
+
> (v1.0) in this English copy.
|
|
8
|
+
|
|
9
|
+
Version: v1.0
|
|
10
|
+
Hierarchy: constitution `HUMAN_AGENT_PARADIGM.md` v1.0 → derived specification `DERIVED_SPECIFICATION.md` v1.0 → this
|
|
11
|
+
template
|
|
12
|
+
Status: execution-layer template · produced with the Owner's authorization
|
|
13
|
+
Applicable scope: P7 delivery and acceptance; submitted as part of the delivery units together with the product,
|
|
14
|
+
evidence package, and reusable assets
|
|
15
|
+
|
|
16
|
+
## Usage Rules
|
|
17
|
+
|
|
18
|
+
1. This report is the index of the evidence package and the contract-comparison statement — **not proof of
|
|
19
|
+
conformance**; conformance rests on independently verifiable evidence (constitution 9.2).
|
|
20
|
+
2. Any "completed/achieved/passed" wording must point to an evidence ID; conclusions without evidence IDs must not
|
|
21
|
+
appear.
|
|
22
|
+
3. Incomplete work, failures, and unmet standards must be listed explicitly; vague wording such as "basically done",
|
|
23
|
+
"almost", or "should work" is forbidden.
|
|
24
|
+
4. The acceptance recommendation is only the Agent's self-check conclusion and is not acceptance evidence; the human
|
|
25
|
+
verifies independently per the contract and then signs.
|
|
26
|
+
5. Before submission, execute《Conformance Self-Check Checklist》v1.0 and attach the results into section 7.
|
|
27
|
+
|
|
28
|
+
---
|
|
29
|
+
|
|
30
|
+
## 1. Metadata
|
|
31
|
+
|
|
32
|
+
| Field | Content |
|
|
33
|
+
|:--------------------------|:--------------------------------|
|
|
34
|
+
| Delivery ID | |
|
|
35
|
+
| Contract ID / version | |
|
|
36
|
+
| Product/task name | |
|
|
37
|
+
| Delivery report version | |
|
|
38
|
+
| Delivery time | |
|
|
39
|
+
| Checklist version | `CONFORMANCE_CHECKLIST.md` v1.0 |
|
|
40
|
+
| Evidence package location | |
|
|
41
|
+
| Product location | |
|
|
42
|
+
| Asset location | |
|
|
43
|
+
|
|
44
|
+
## 2. Delivery Units List
|
|
45
|
+
|
|
46
|
+
| Category | Content | Location/link |
|
|
47
|
+
|:-----------------|:------------|:--------------|
|
|
48
|
+
| Product itself | | |
|
|
49
|
+
| Evidence package | | |
|
|
50
|
+
| Reusable assets | | |
|
|
51
|
+
| Delivery report | this report | |
|
|
52
|
+
|
|
53
|
+
## 3. Contract Comparison Master Table
|
|
54
|
+
|
|
55
|
+
| Contract clause ID | Contract criterion/deliverable | Evidence ID | Review method | Conclusion |
|
|
56
|
+
|:-------------------|:-------------------------------|:------------|:--------------|:------------------|
|
|
57
|
+
| C2-S1 | | | | `Met` / `Not met` |
|
|
58
|
+
| C4-1 | | | | |
|
|
59
|
+
| C5-E6.x | | | | |
|
|
60
|
+
|
|
61
|
+
**Rules:** conclusions may only be `Met` / `Not met`; `Met` must be checkable by a third party from the evidence.
|
|
62
|
+
|
|
63
|
+
## 4. Reliability Evidence
|
|
64
|
+
|
|
65
|
+
### 4.1 Completion-claim mapping
|
|
66
|
+
|
|
67
|
+
| Claim | Corresponding contract clause | Evidence ID | Evidence type |
|
|
68
|
+
|:------|:------------------------------|:------------|:---------------------------|
|
|
69
|
+
| | | | run/use/test records, etc. |
|
|
70
|
+
|
|
71
|
+
### 4.2 State consistency
|
|
72
|
+
|
|
73
|
+
| Checkpoint | Externally reported state | Real-state snapshot evidence | Consistent? |
|
|
74
|
+
|:-----------|:--------------------------|:-----------------------------|:------------|
|
|
75
|
+
| | | | |
|
|
76
|
+
|
|
77
|
+
### 4.3 Traceability chain
|
|
78
|
+
|
|
79
|
+
| Need | Design | Implementation | Verification | Delivery |
|
|
80
|
+
|:-----|:-------|:---------------|:-------------|:---------|
|
|
81
|
+
| | | | | |
|
|
82
|
+
|
|
83
|
+
### 4.4 Failures and incomplete items
|
|
84
|
+
|
|
85
|
+
| No. | Failure/incomplete item | Impact | Handling | Evidence ID |
|
|
86
|
+
|:----|:------------------------|:-------|:---------|:------------|
|
|
87
|
+
| | | | | |
|
|
88
|
+
|
|
89
|
+
**Statement:** none of the incomplete items above is presented as complete elsewhere in this delivery report.
|
|
90
|
+
|
|
91
|
+
## 5. Excellence Evidence (per dimension)
|
|
92
|
+
|
|
93
|
+
| Dimension | Contract criterion ID | Judgment | Evidence ID | Third-party review method |
|
|
94
|
+
|:---------------------------|:----------------------|:------------------|:------------|:--------------------------|
|
|
95
|
+
| 6.1 Sound design | | `Met` / `Not met` | | |
|
|
96
|
+
| 6.2 Complete functionality | | | | |
|
|
97
|
+
| 6.3 Elegant implementation | | | | |
|
|
98
|
+
| 6.4 Perfect experience | | | | |
|
|
99
|
+
| 6.5 Excellent performance | | | | |
|
|
100
|
+
|
|
101
|
+
**Rules:** none of the five dimensions may be missing; outstanding performance on one dimension does not offset failure
|
|
102
|
+
on another.
|
|
103
|
+
|
|
104
|
+
## 6. Multi-View Review and Defects
|
|
105
|
+
|
|
106
|
+
### 6.1 Review records
|
|
107
|
+
|
|
108
|
+
| Perspective | Review standpoint | Defect-finding method | Counterexamples/paths attempted | Problems found | Record ID |
|
|
109
|
+
|:------------|:------------------|:----------------------|:--------------------------------|:---------------|:----------|
|
|
110
|
+
| Product | | | | | |
|
|
111
|
+
| User | | | | | |
|
|
112
|
+
| Engineering | | | | | |
|
|
113
|
+
| Adversarial | | | | | |
|
|
114
|
+
|
|
115
|
+
### 6.2 Defect list
|
|
116
|
+
|
|
117
|
+
| Defect ID | Severity | In/out of contract scope | Handling | Re-verification evidence ID | Status |
|
|
118
|
+
|:----------|:--------------------|:-------------------------|:---------|:----------------------------|:--------------------------------|
|
|
119
|
+
| | Blocker/Major/Minor | | | | `Closed` / `Moved out of scope` |
|
|
120
|
+
|
|
121
|
+
**Exit judgment:** all in-contract defects closed; defects moved out of scope were human-approved and are explicitly
|
|
122
|
+
disclosed here (never silently hidden).
|
|
123
|
+
|
|
124
|
+
## 7. Conformance Self-Check Results
|
|
125
|
+
|
|
126
|
+
| Item | Result |
|
|
127
|
+
|:-------------------------------------------------|:-------|
|
|
128
|
+
| Checklist version | |
|
|
129
|
+
| Groups A–I items total / passed / not applicable | |
|
|
130
|
+
| ★ hard items total / passed | |
|
|
131
|
+
| Non-conformances and rectification evidence | |
|
|
132
|
+
| Paradigm-level group J (applicable / passed) | |
|
|
133
|
+
|
|
134
|
+
**Statement:** passing self-check is only the Agent's self-check conclusion and is not proof of conformance; final
|
|
135
|
+
conformance rests on behavior and results, evidenced by independently verifiable records.
|
|
136
|
+
|
|
137
|
+
## 8. Asset Deposition
|
|
138
|
+
|
|
139
|
+
| Asset ID | Type | Source | Verification evidence | Owner | Usage authorization scope |
|
|
140
|
+
|:---------|:--------------------------------------|:-------|:----------------------|:----------|:--------------------------|
|
|
141
|
+
| | preference/criterion/component/lesson | | | the human | |
|
|
142
|
+
|
|
143
|
+
## 9. Cost Ledger Summary
|
|
144
|
+
|
|
145
|
+
| Cost item | Budget | Actual | Variance note |
|
|
146
|
+
|:-------------------------------------------|:-------|:-------|:--------------|
|
|
147
|
+
| Human attention (decisions/waiting/rework) | | | |
|
|
148
|
+
| Agent consumption | | | |
|
|
149
|
+
| Rework | | | |
|
|
150
|
+
| Error and delay costs | | | |
|
|
151
|
+
|
|
152
|
+
## 10. Safety and Compliance
|
|
153
|
+
|
|
154
|
+
| Check item | Result | Evidence ID |
|
|
155
|
+
|:--------------------------------------------|:-------|:------------|
|
|
156
|
+
| Safety/compliance risk identification | | |
|
|
157
|
+
| Risk handling/escalation | | |
|
|
158
|
+
| Bottom-line violation record (must be zero) | | |
|
|
159
|
+
| Residual-risk disclosure | | |
|
|
160
|
+
|
|
161
|
+
## 11. Known Boundaries and Incomplete-Item Statement
|
|
162
|
+
|
|
163
|
+
**List all boundaries, incomplete items, and moved-out-of-scope items explicitly — no hiding or blurring:**
|
|
164
|
+
|
|
165
|
+
## 12. Acceptance Recommendation (self-check conclusion only)
|
|
166
|
+
|
|
167
|
+
| Item | Content |
|
|
168
|
+
|:-----------------------|:---------------------------------------------------------------------------------------------|
|
|
169
|
+
| Recommended conclusion | `Recommend acceptance` / `Do not recommend acceptance` |
|
|
170
|
+
| Basis | item-by-item comparison results of sections 3, 4, 5, 6, 7 |
|
|
171
|
+
| Nature statement | this recommendation is only the Agent's self-check conclusion and is not acceptance evidence |
|
|
172
|
+
|
|
173
|
+
## 13. Signature and Acceptance
|
|
174
|
+
|
|
175
|
+
| Role | Action | Signature | Time |
|
|
176
|
+
|:-------------------------|:-------------------------------------------------------------------------------------------|:----------|:-----|
|
|
177
|
+
| Agent delivering | warrant the report content matches the evidence package and nothing un-happened is claimed | | |
|
|
178
|
+
| Independent review | re-check section 3 matrix against the evidence package | | |
|
|
179
|
+
| Human (Owner) acceptance | accept and sign the conclusion per the contract | | |
|
|
180
|
+
|
|
181
|
+
**Acceptance conclusion:** `Pass` / `Fail`
|
|
182
|
+
**Acceptance basis and time:**
|
|
183
|
+
|
|
184
|
+
## Appendix: Traceability Matrix
|
|
185
|
+
|
|
186
|
+
| Constitutional clause | Derived spec | Contract clause | Evidence ID | Conclusion |
|
|
187
|
+
|:----------------------|:-------------|:----------------|:------------|:-----------|
|
|
188
|
+
| | | | | |
|