Skip to content

Commit a8b3734

Browse files
chaokunyangchaokunyang
committed
add ai contributing policy
1 parent 550d8ee commit a8b3734

2 files changed

Lines changed: 138 additions & 0 deletions

File tree

‎.github/pull_request_template.md‎

Lines changed: 11 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -28,6 +28,17 @@ Is there any related issue? If this PR closes them you say say fix/closes:
2828
- Fixes #xxxx2
2929
-->
3030

31+
## AI Contribution Checklist (required when AI assistance = `yes`)
32+
33+
- [ ] AI assistance was used in this PR: `yes` / `no`
34+
- [ ] I can explain and defend all important changes without AI help.
35+
- [ ] I reviewed AI-assisted code changes line by line before submission.
36+
- [ ] I summarized AI assistance scope in `What does this PR do?`.
37+
- [ ] I listed affected files/subsystems in `What does this PR do?`.
38+
- [ ] I recorded human verification evidence (commands/tests/benchmarks and pass/fail summary) in this PR.
39+
- [ ] I added/updated tests and specs where required.
40+
- [ ] I validated protocol/performance impacts with evidence when applicable.
41+
3142
## Does this PR introduce any user-facing change?
3243

3344
<!--

‎AI_CONTRIBUTION_POLICY.md‎

Lines changed: 127 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,127 @@
1+
# AI Contribution Policy
2+
3+
Apache Fory is a performance-critical foundational serialization framework with cross-language compatibility requirements.
4+
AI tools are welcome as assistants, but project quality, legal safety, and maintainability standards are unchanged.
5+
6+
The key words MUST, MUST NOT, REQUIRED, SHOULD, and MAY are interpreted as described in RFC 2119.
7+
8+
## 1. Core Principle
9+
10+
- AI tools MAY assist contribution work.
11+
- AI tools MUST NOT replace contributor accountability.
12+
- The human submitter is responsible for correctness, safety, performance, and maintainability of all submitted changes.
13+
- AI-assisted code MUST be reviewed carefully by the contributor line by line before submission.
14+
- Contributors MUST be able to explain and defend design and implementation details during review.
15+
16+
## 2. Disclosure (Privacy-Safe)
17+
18+
For substantial AI assistance, PR descriptions MUST include a short `AI Usage Disclosure` section.
19+
20+
Required disclosure fields:
21+
22+
- Whether substantial AI assistance was used (`yes` or `no`)
23+
- Scope of assistance (for example: design drafting, code drafting, refactor suggestions, tests, docs)
24+
- Affected files or subsystems (high-level)
25+
- Human verification performed (commands/tests/benchmarks executed)
26+
- Provenance and license confirmation (see Section 6)
27+
28+
To protect privacy and enterprise security:
29+
30+
- Contributors are NOT required to disclose model names, provider names, private prompts, or internal workflow details.
31+
- Maintainers MAY request additional clarification only when necessary for legal or technical review.
32+
33+
## 3. Human Communication Requirements
34+
35+
The following MUST be human-authored (translation and grammar correction tools are acceptable):
36+
37+
- Review responses
38+
- Design rationale and tradeoff discussion
39+
- Risk analysis and production impact explanation
40+
41+
Generated filler text, evasive responses, or content that does not reflect contributor understanding may result in PR closure.
42+
43+
## 4. Scope Alignment Before Implementation
44+
45+
For non-trivial changes (especially architecture/protocol/performance-sensitive work), contributors SHOULD align scope in an issue or discussion before implementation.
46+
47+
This expectation applies to all contributors, whether AI-assisted or not.
48+
For AI-assisted non-trivial work without prior alignment, maintainers MAY request scope alignment before continuing review.
49+
50+
## 5. Verification Requirements
51+
52+
Every AI-assisted PR MUST provide verifiable evidence of local validation:
53+
54+
- Confirmation that contributor performed line-by-line self-review of AI-assisted code changes
55+
- Build/lint/test commands run locally
56+
- Targeted tests for changed behavior
57+
- Results summary (pass/fail and relevant environment context)
58+
59+
Additional REQUIRED checks for Fory-critical paths:
60+
61+
- Protocol, type mapping, or wire-format changes:
62+
- Update relevant docs under `docs/specification/**`
63+
- Add or update cross-language compatibility tests where applicable
64+
- Performance-sensitive changes (serialization/deserialization hot paths, memory allocation behavior, codegen, buffer logic):
65+
- Provide benchmark or regression evidence
66+
- Justify any measurable performance or allocation impact
67+
68+
Claims without evidence may be treated as incomplete.
69+
70+
## 6. Licensing, Copyright, and Provenance
71+
72+
Contributors MUST follow ASF legal guidance and project licensing policy, including:
73+
74+
- ASF Generative Tooling Guidance
75+
- ASF third-party licensing requirements
76+
- Apache-2.0 compatibility obligations
77+
78+
Contributors MUST ensure:
79+
80+
- No incompatible third-party code is introduced
81+
- Reused material has compatible licensing and required attribution
82+
- AI-generated output does not include unauthorized copyrighted fragments
83+
84+
If provenance is uncertain, contributors MUST remove or replace the material before submission.
85+
Maintainers MAY request provenance clarification when needed.
86+
87+
## 7. Quality Gate and Non-Acceptance Conditions
88+
89+
Maintainers MAY close or return PRs that materially fail project standards, including:
90+
91+
- Contributor cannot explain key implementation logic
92+
- Missing required disclosure for substantial AI assistance
93+
- Missing verification evidence for changed behavior
94+
- Redundant implementation of existing utilities without clear necessity
95+
- Introduction of dead code, unused helpers, or placeholder abstractions without justification
96+
- Protocol or performance claims without reproducible evidence
97+
- Large unfocused changes with unclear scope or ownership
98+
99+
This is not a ban on AI usage; it is a quality and maintainability gate.
100+
101+
## 8. Review and Enforcement Process
102+
103+
Before merge, maintainers MAY request:
104+
105+
- Additional tests, benchmarks, or spec updates
106+
- PR split into smaller verifiable commits
107+
- Clarification of technical rationale, provenance, or licensing
108+
- Rework of sections that do not meet standards
109+
110+
Maintainers MAY close PRs that remain non-compliant after feedback.
111+
112+
Any long-term contribution restrictions MUST follow Apache project governance and community process, and SHOULD be documented with clear rationale.
113+
114+
## 9. Contributor Checklist (for AI-Assisted PRs)
115+
116+
- [ ] I can explain and defend all important changes.
117+
- [ ] I reviewed AI-assisted code line by line before submission.
118+
- [ ] I provided `AI Usage Disclosure` without exposing private/internal details.
119+
- [ ] I ran relevant local build/lint/test checks and reported outcomes.
120+
- [ ] I added/updated tests and specs where required.
121+
- [ ] I validated protocol/performance impacts with evidence when applicable.
122+
- [ ] I verified licensing and provenance compliance.
123+
124+
## 10. Governance Note
125+
126+
This policy complements, but does not replace, existing ASF and Apache Fory governance, contribution, and legal policies.
127+
If conflicts arise, ASF legal and project governance rules take precedence.

0 commit comments

Comments
 (0)