feat(skills): add research assistant skill suite
This commit is contained in:
NeeNe
parent
9749a4c832
commit
8642f4026e
|
|
@ -26,5 +26,8 @@ jobs:
|
|||
- name: Test
|
||||
run: make test
|
||||
|
||||
- name: Validate research Skills
|
||||
run: make validate-research-skills
|
||||
|
||||
- name: Check formatting
|
||||
run: make fmt
|
||||
|
|
|
|||
|
|
@ -26,5 +26,8 @@ jobs:
|
|||
- name: Scan i18n key references
|
||||
run: go run ./internal/i18n/cmd/check --scan-code
|
||||
|
||||
- name: Validate research Skills
|
||||
run: python3 scripts/validate-research-skills.py
|
||||
|
||||
- name: Run Go tests
|
||||
run: go test ./...
|
||||
|
|
|
|||
7
Makefile
7
Makefile
|
|
@ -3,7 +3,7 @@ BINARY := gitlink-cli
|
|||
VERSION ?= $(shell git describe --tags --always --dirty 2>/dev/null || echo "dev")
|
||||
LDFLAGS := -s -w -X '$(MODULE)/cmd.Version=$(VERSION)'
|
||||
|
||||
.PHONY: build install clean test check vet fmt cover lint
|
||||
.PHONY: build install clean test check vet fmt cover lint validate-research-skills
|
||||
|
||||
build:
|
||||
go build -ldflags "$(LDFLAGS)" -o $(BINARY) .
|
||||
|
|
@ -35,7 +35,10 @@ cover:
|
|||
lint:
|
||||
golangci-lint run ./...
|
||||
|
||||
check: fmt vet lint test
|
||||
validate-research-skills:
|
||||
python3 scripts/validate-research-skills.py
|
||||
|
||||
check: fmt vet lint test validate-research-skills
|
||||
@echo "All checks passed."
|
||||
|
||||
hooks:
|
||||
|
|
|
|||
|
|
@ -0,0 +1,22 @@
|
|||
# Research skills suite
|
||||
|
||||
Adds seven read-only research-oriented Agent Skills for GitLink repositories:
|
||||
|
||||
- `gitlink-research-reproducibility`: reproducibility checklist and remediation plan for research code.
|
||||
- `gitlink-research-compliance`: license, security policy, dependency, and sensitive-file risk audit.
|
||||
- `gitlink-research-progress-tracker`: weekly progress report and risk early warning for research projects.
|
||||
- `gitlink-research-collaboration-map`: contributor, Issue, and PR collaboration mapping for research teams.
|
||||
- `gitlink-research-knowledge-graph`: keyword-based research landscape and lightweight knowledge graph workflow.
|
||||
- `gitlink-research-data-provenance`: dataset source, citation, license, and privacy-risk audit.
|
||||
- `gitlink-research-artifact-handbook`: research artifact handbook generation for demos, handoff, and open-source release.
|
||||
|
||||
These Skills are designed as standalone `gitlink-cli` Skill PR deliverables. They do not depend on external project code and can also be composed by end-to-end research workflows.
|
||||
|
||||
Validation:
|
||||
|
||||
```bash
|
||||
make validate-research-skills
|
||||
git diff --check
|
||||
```
|
||||
|
||||
Agent verification guidance is documented in `docs/research-skills-agent-test-report.md`.
|
||||
|
|
@ -0,0 +1,147 @@
|
|||
# Research Skills Agent Test Report
|
||||
|
||||
## Scope
|
||||
|
||||
This report covers the seven read-only research Skills added for the GitLink competition:
|
||||
|
||||
- `gitlink-research-reproducibility`
|
||||
- `gitlink-research-compliance`
|
||||
- `gitlink-research-progress-tracker`
|
||||
- `gitlink-research-collaboration-map`
|
||||
- `gitlink-research-knowledge-graph`
|
||||
- `gitlink-research-data-provenance`
|
||||
- `gitlink-research-artifact-handbook`
|
||||
|
||||
The Skills are standalone `gitlink-cli` Agent Skills. They do not depend on external project code and can be submitted as an independent Skill PR.
|
||||
|
||||
## Automatic Validation
|
||||
|
||||
Run:
|
||||
|
||||
```bash
|
||||
make validate-research-skills
|
||||
git diff --check
|
||||
```
|
||||
|
||||
The validation script checks:
|
||||
|
||||
- every Skill has valid `SKILL.md` frontmatter with the expected name and a useful trigger description;
|
||||
- every Skill instructs the Agent to read `gitlink-shared`;
|
||||
- every Skill is read-only by default and uses `gitlink-cli --format json`;
|
||||
- every Skill includes command coverage for its scenario;
|
||||
- every Skill has at least one reference file and one example file;
|
||||
- examples contain user request, Agent steps, expected answer, and concrete `gitlink-cli` commands;
|
||||
- `skills/README.md` and `doc/changes/research-skills-suite.md` index all seven Skills.
|
||||
|
||||
## Offline Agent Test Prompts
|
||||
|
||||
These prompts can be used in Claude Code, Cursor, Codex, or OpenClaw after installing the Skills.
|
||||
|
||||
### Reproducibility
|
||||
|
||||
```text
|
||||
Use gitlink-research-reproducibility to audit songhui18/ICCV2021. Produce a reproducibility score, missing items, and top three remediation actions. Do not write back to GitLink.
|
||||
```
|
||||
|
||||
Expected behavior:
|
||||
|
||||
- read `gitlink-shared`;
|
||||
- run `repo +info` and `repo +tree`, or fall back to `api GET /sub_entries`;
|
||||
- cite command sources;
|
||||
- produce a checklist report.
|
||||
|
||||
### Compliance
|
||||
|
||||
```text
|
||||
Use gitlink-research-compliance to check Gitconomy/Git4Research before open-source release. Report license, SECURITY, CONTRIBUTING, dependency, and sensitive-file risks.
|
||||
```
|
||||
|
||||
Expected behavior:
|
||||
|
||||
- perform read-only file tree checks;
|
||||
- avoid printing sensitive contents;
|
||||
- state that the output is an engineering pre-check, not legal advice.
|
||||
|
||||
### Progress Tracking
|
||||
|
||||
```text
|
||||
Use gitlink-research-progress-tracker to generate a weekly research progress report for songhui18/ICCV2021, including Issue/PR queues and risk warnings.
|
||||
```
|
||||
|
||||
Expected behavior:
|
||||
|
||||
- collect repo, Issue, PR, Release, and contributor data;
|
||||
- separate facts from recommendations;
|
||||
- avoid posting a comment unless explicitly requested.
|
||||
|
||||
### Collaboration Map
|
||||
|
||||
```text
|
||||
Use gitlink-research-collaboration-map to analyze collaboration structure for Gitconomy/Git4Research. Identify maintainers, contributors, open collaboration gaps, and suggested next actions.
|
||||
```
|
||||
|
||||
Expected behavior:
|
||||
|
||||
- use public contributor, Issue, PR, language, and optional user data;
|
||||
- avoid private contact information;
|
||||
- base recommendations on evidence.
|
||||
|
||||
### Knowledge Graph
|
||||
|
||||
```text
|
||||
Use gitlink-research-knowledge-graph to search GitLink for “论文复现” and “open research”, then output a Markdown trend summary and graph JSON.
|
||||
```
|
||||
|
||||
Expected behavior:
|
||||
|
||||
- run bounded keyword searches;
|
||||
- deep-analyze no more than eight repositories;
|
||||
- produce `nodes` and `edges` using the documented graph schema;
|
||||
- mark platform-data limits.
|
||||
|
||||
### Data Provenance
|
||||
|
||||
```text
|
||||
Use gitlink-research-data-provenance to audit songhui18/ICCV2021 for dataset source, citation, license, and privacy risks. Do not print raw data contents.
|
||||
```
|
||||
|
||||
Expected behavior:
|
||||
|
||||
- collect repo info and selected file trees;
|
||||
- report paths and risk types only;
|
||||
- distinguish missing evidence from confirmed risk.
|
||||
|
||||
### Artifact Handbook
|
||||
|
||||
```text
|
||||
Use gitlink-research-artifact-handbook to generate a research artifact handbook for Gitconomy/Git4Research, suitable for project handoff and competition demo.
|
||||
```
|
||||
|
||||
Expected behavior:
|
||||
|
||||
- collect repo, README, tree, Issue/PR, Release, language, and contributor signals when available;
|
||||
- produce a structured handbook with missing fields marked as `待补充`;
|
||||
- avoid writing Wiki, Issue, Release, or repository files unless explicitly requested.
|
||||
|
||||
## Real Agent Verification
|
||||
|
||||
For competition submission, record at least one Agent run with screenshots or video:
|
||||
|
||||
1. Install or expose the Skills to the Agent.
|
||||
2. Run one prompt from this report against a public GitLink repository.
|
||||
3. Show the Agent reading the Skill, running read-only `gitlink-cli` commands, and producing the report.
|
||||
4. Show that no write operation was performed.
|
||||
|
||||
Recommended demo target:
|
||||
|
||||
```text
|
||||
songhui18/ICCV2021
|
||||
```
|
||||
|
||||
This repository has richer Issue and PR data, which makes progress tracking and collaboration mapping more visible.
|
||||
|
||||
## Known Limitations
|
||||
|
||||
- The automatic script validates Skill structure and workflow instructions; it does not prove semantic quality of an LLM-generated report.
|
||||
- Real GitLink API verification requires network access and may require authentication for private repositories.
|
||||
- Screenshots or video still need to be captured in the selected Agent UI for the competition submission package.
|
||||
|
|
@ -0,0 +1,160 @@
|
|||
#!/usr/bin/env python3
|
||||
from __future__ import annotations
|
||||
|
||||
import re
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
|
||||
RESEARCH_SKILLS = {
|
||||
"gitlink-research-reproducibility": {
|
||||
"commands": ["repo +info", "repo +tree", "api GET"],
|
||||
"references": ["checklist.md"],
|
||||
"examples": ["reproducibility-audit.md"],
|
||||
"keywords": ["复现", "README", "LICENSE", "依赖"],
|
||||
},
|
||||
"gitlink-research-compliance": {
|
||||
"commands": ["repo +info", "repo +tree", "api GET"],
|
||||
"references": ["risk-rules.md"],
|
||||
"examples": ["compliance-audit.md"],
|
||||
"keywords": ["合规", "敏感", "LICENSE", "SECURITY"],
|
||||
},
|
||||
"gitlink-research-progress-tracker": {
|
||||
"commands": ["issue +list", "pr +list", "release +list"],
|
||||
"references": ["risk-model.md"],
|
||||
"examples": ["weekly-report.md"],
|
||||
"keywords": ["进度", "周报", "风险", "预警"],
|
||||
},
|
||||
"gitlink-research-collaboration-map": {
|
||||
"commands": ["repo +contributors", "issue +list", "pr +list"],
|
||||
"references": ["profile-fields.md"],
|
||||
"examples": ["collaboration-map.md"],
|
||||
"keywords": ["协作", "贡献者", "合作", "画像"],
|
||||
},
|
||||
"gitlink-research-knowledge-graph": {
|
||||
"commands": ["search +repos", "repo +info", "repo +languages"],
|
||||
"references": ["graph-schema.md"],
|
||||
"examples": ["keyword-map.md"],
|
||||
"keywords": ["知识图谱", "热点", "选题", "趋势"],
|
||||
},
|
||||
"gitlink-research-data-provenance": {
|
||||
"commands": ["repo +info", "repo +tree", "api GET"],
|
||||
"references": ["provenance-rules.md"],
|
||||
"examples": ["data-provenance-audit.md"],
|
||||
"keywords": ["数据", "隐私", "来源", "许可证"],
|
||||
},
|
||||
"gitlink-research-artifact-handbook": {
|
||||
"commands": ["repo +info", "repo +tree", "repo +readme", "release +list"],
|
||||
"references": ["handbook-template.md"],
|
||||
"examples": ["artifact-handbook.md"],
|
||||
"keywords": ["成果", "手册", "交接", "复现"],
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
def fail(message: str) -> None:
|
||||
print(f"ERROR: {message}", file=sys.stderr)
|
||||
raise SystemExit(1)
|
||||
|
||||
|
||||
def read(path: Path) -> str:
|
||||
try:
|
||||
return path.read_text(encoding="utf-8")
|
||||
except FileNotFoundError:
|
||||
fail(f"missing file: {path}")
|
||||
|
||||
|
||||
def parse_frontmatter(text: str, path: Path) -> dict[str, str]:
|
||||
if not text.startswith("---\n"):
|
||||
fail(f"{path} does not start with YAML frontmatter")
|
||||
try:
|
||||
_, frontmatter, _ = text.split("---", 2)
|
||||
except ValueError:
|
||||
fail(f"{path} has incomplete YAML frontmatter")
|
||||
parsed: dict[str, str] = {}
|
||||
for line in frontmatter.strip().splitlines():
|
||||
if ":" not in line:
|
||||
fail(f"{path} frontmatter line has no colon: {line}")
|
||||
key, value = line.split(":", 1)
|
||||
parsed[key.strip()] = value.strip().strip('"')
|
||||
return parsed
|
||||
|
||||
|
||||
def assert_contains(text: str, needle: str, path: Path) -> None:
|
||||
if needle not in text:
|
||||
fail(f"{path} is missing required text: {needle}")
|
||||
|
||||
|
||||
def assert_regex(text: str, pattern: str, path: Path) -> None:
|
||||
if not re.search(pattern, text, flags=re.MULTILINE):
|
||||
fail(f"{path} does not match required pattern: {pattern}")
|
||||
|
||||
|
||||
def validate_skill(root: Path, name: str, spec: dict[str, list[str]]) -> None:
|
||||
skill_dir = root / "skills" / name
|
||||
skill_md = skill_dir / "SKILL.md"
|
||||
text = read(skill_md)
|
||||
frontmatter = parse_frontmatter(text, skill_md)
|
||||
|
||||
if frontmatter.get("name") != name:
|
||||
fail(f"{skill_md} has wrong name: {frontmatter.get('name')}")
|
||||
description = frontmatter.get("description", "")
|
||||
if len(description) < 30:
|
||||
fail(f"{skill_md} description is too short")
|
||||
for keyword in spec["keywords"]:
|
||||
if keyword not in text and keyword not in description:
|
||||
fail(f"{skill_md} is missing scenario keyword: {keyword}")
|
||||
|
||||
assert_contains(text, "gitlink-shared", skill_md)
|
||||
assert_contains(text, "gitlink-cli", skill_md)
|
||||
assert_contains(text, "--format json", skill_md)
|
||||
assert_regex(text, r"只读|默认只读", skill_md)
|
||||
assert_regex(text, r"不输出 Token|Token", skill_md)
|
||||
|
||||
for command in spec["commands"]:
|
||||
assert_contains(text, command, skill_md)
|
||||
|
||||
for filename in spec["references"]:
|
||||
path = skill_dir / "references" / filename
|
||||
ref_text = read(path)
|
||||
if len(ref_text.strip().splitlines()) < 5:
|
||||
fail(f"{path} is too small to be useful")
|
||||
|
||||
for filename in spec["examples"]:
|
||||
path = skill_dir / "examples" / filename
|
||||
example_text = read(path)
|
||||
assert_contains(example_text, "User request", path)
|
||||
assert_contains(example_text, "Agent steps", path)
|
||||
assert_contains(example_text, "Expected answer", path)
|
||||
assert_contains(example_text, "gitlink-cli", path)
|
||||
|
||||
|
||||
def validate_readme(root: Path) -> None:
|
||||
readme = root / "skills" / "README.md"
|
||||
text = read(readme)
|
||||
for name in RESEARCH_SKILLS:
|
||||
assert_contains(text, name, readme)
|
||||
|
||||
|
||||
def validate_change_note(root: Path) -> None:
|
||||
note = root / "doc" / "changes" / "research-skills-suite.md"
|
||||
text = read(note)
|
||||
for name in RESEARCH_SKILLS:
|
||||
assert_contains(text, name, note)
|
||||
if "scripts/validate-research-skills.py" not in text and "make validate-research-skills" not in text:
|
||||
fail(f"{note} is missing the research Skills validation command")
|
||||
|
||||
|
||||
def main() -> int:
|
||||
root = Path(__file__).resolve().parents[1]
|
||||
for name, spec in RESEARCH_SKILLS.items():
|
||||
validate_skill(root, name, spec)
|
||||
print(f"ok {name}")
|
||||
validate_readme(root)
|
||||
validate_change_note(root)
|
||||
print("research skill validation passed")
|
||||
return 0
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
raise SystemExit(main())
|
||||
|
|
@ -105,6 +105,48 @@ skills/
|
|||
│ └── SKILL.md # Wiki 操作指南
|
||||
├── gitlink-pm/ # 项目管理
|
||||
│ └── SKILL.md # PM 操作指南
|
||||
├── gitlink-research-reproducibility/ # 科研仓库复现性审计
|
||||
│ ├── SKILL.md
|
||||
│ ├── references/
|
||||
│ │ └── checklist.md
|
||||
│ └── examples/
|
||||
│ └── reproducibility-audit.md
|
||||
├── gitlink-research-compliance/ # 科研开源合规与敏感风险检查
|
||||
│ ├── SKILL.md
|
||||
│ ├── references/
|
||||
│ │ └── risk-rules.md
|
||||
│ └── examples/
|
||||
│ └── compliance-audit.md
|
||||
├── gitlink-research-progress-tracker/ # 科研项目进度跟踪与预警
|
||||
│ ├── SKILL.md
|
||||
│ ├── references/
|
||||
│ │ └── risk-model.md
|
||||
│ └── examples/
|
||||
│ └── weekly-report.md
|
||||
├── gitlink-research-collaboration-map/ # 科研协作画像与合作建议
|
||||
│ ├── SKILL.md
|
||||
│ ├── references/
|
||||
│ │ └── profile-fields.md
|
||||
│ └── examples/
|
||||
│ └── collaboration-map.md
|
||||
├── gitlink-research-knowledge-graph/ # 科研热点追踪与轻量知识图谱
|
||||
│ ├── SKILL.md
|
||||
│ ├── references/
|
||||
│ │ └── graph-schema.md
|
||||
│ └── examples/
|
||||
│ └── keyword-map.md
|
||||
├── gitlink-research-data-provenance/ # 科研数据来源与隐私审计
|
||||
│ ├── SKILL.md
|
||||
│ ├── references/
|
||||
│ │ └── provenance-rules.md
|
||||
│ └── examples/
|
||||
│ └── data-provenance-audit.md
|
||||
├── gitlink-research-artifact-handbook/ # 科研成果沉淀手册生成
|
||||
│ ├── SKILL.md
|
||||
│ ├── references/
|
||||
│ │ └── handbook-template.md
|
||||
│ └── examples/
|
||||
│ └── artifact-handbook.md
|
||||
├── gitlink-health/ # 项目健康度分析
|
||||
│ ├── SKILL.md # 健康度分析指南
|
||||
│ ├── data/
|
||||
|
|
@ -148,6 +190,18 @@ skills/
|
|||
| **gitlink-workflow** | AI 工作流 | Issue 分类、PR Review、Release Notes |
|
||||
| **gitlink-health** | 开源项目健康度 | 详情见SKILL.md |
|
||||
|
||||
### 科研场景 Skills
|
||||
|
||||
| Skill | 说明 | 常用命令 |
|
||||
|-------|------|----------|
|
||||
| **gitlink-research-reproducibility** | 科研仓库复现性审计，检查 README、LICENSE、依赖、测试、CI、示例和数据说明 | `repo +info`, `repo +tree`, `api GET /sub_entries` |
|
||||
| **gitlink-research-compliance** | 科研开源合规与敏感风险检查 | `repo +info`, `repo +tree`, `api GET /sub_entries` |
|
||||
| **gitlink-research-progress-tracker** | 科研项目周报、Issue/PR 队列和停滞风险预警 | `issue +list`, `pr +list`, `release +list`, `repo +contributors` |
|
||||
| **gitlink-research-collaboration-map** | 科研协作画像、核心维护者识别和合作建议 | `repo +contributors`, `issue +list`, `pr +list`, `user +info` |
|
||||
| **gitlink-research-knowledge-graph** | 关键词科研热点追踪和轻量知识图谱 | `search +repos`, `repo +info`, `repo +languages`, `repo +contributors` |
|
||||
| **gitlink-research-data-provenance** | 科研数据来源、数据许可证、引用链和隐私风险审计 | `repo +info`, `repo +tree`, `api GET /sub_entries` |
|
||||
| **gitlink-research-artifact-handbook** | 科研成果沉淀手册和课题组交接材料生成 | `repo +info`, `repo +readme`, `repo +tree`, `release +list` |
|
||||
|
||||
---
|
||||
|
||||
## 🎯 使用场景
|
||||
|
|
@ -252,6 +306,15 @@ gitlink-cli org +info -i Gitlink
|
|||
- [gitlink-org/SKILL.md](gitlink-org/SKILL.md) - 组织命令
|
||||
- [gitlink-user/SKILL.md](gitlink-user/SKILL.md) - 用户命令
|
||||
|
||||
**科研场景**:
|
||||
- [gitlink-research-reproducibility/SKILL.md](gitlink-research-reproducibility/SKILL.md) - 复现性审计
|
||||
- [gitlink-research-compliance/SKILL.md](gitlink-research-compliance/SKILL.md) - 开源合规与敏感风险
|
||||
- [gitlink-research-progress-tracker/SKILL.md](gitlink-research-progress-tracker/SKILL.md) - 项目进度跟踪与预警
|
||||
- [gitlink-research-collaboration-map/SKILL.md](gitlink-research-collaboration-map/SKILL.md) - 协作画像与合作建议
|
||||
- [gitlink-research-knowledge-graph/SKILL.md](gitlink-research-knowledge-graph/SKILL.md) - 热点追踪与知识图谱
|
||||
- [gitlink-research-data-provenance/SKILL.md](gitlink-research-data-provenance/SKILL.md) - 数据来源与隐私审计
|
||||
- [gitlink-research-artifact-handbook/SKILL.md](gitlink-research-artifact-handbook/SKILL.md) - 成果沉淀手册
|
||||
|
||||
---
|
||||
|
||||
## ❓ 常见问题
|
||||
|
|
|
|||
|
|
@ -0,0 +1,73 @@
|
|||
---
|
||||
name: gitlink-research-artifact-handbook
|
||||
description: "科研成果沉淀手册生成：基于 GitLink 仓库 README、目录结构、Release、Issue、PR、贡献者和语言信息，生成面向答辩、开源发布或课题组交接的科研成果手册。用于梳理代码、实验、论文、数据、贡献和复现入口。"
|
||||
---
|
||||
|
||||
# gitlink-research-artifact-handbook
|
||||
|
||||
开始前必须先阅读 `../gitlink-shared/SKILL.md`，确认认证、权限、安全规则和 GitLink API 注意事项。
|
||||
|
||||
## 安全规则
|
||||
|
||||
- 默认只读执行，不创建 Wiki、不发布 Release、不提交文件。
|
||||
- 所有 GitLink 操作必须使用 `gitlink-cli`。
|
||||
- 所有命令使用 `--format json`。
|
||||
- 不输出 Token、Cookie 或认证 Header。
|
||||
- 如果用户要求把手册写回 Wiki、Issue 或仓库文件，必须先输出 dry-run 草稿并等待确认。
|
||||
|
||||
## 工作流
|
||||
|
||||
1. 采集仓库概览、语言和贡献者：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner <owner> --repo <repo> --format json
|
||||
gitlink-cli repo +languages --owner <owner> --repo <repo> --format json
|
||||
gitlink-cli repo +contributors --owner <owner> --repo <repo> --format json
|
||||
```
|
||||
|
||||
2. 采集根目录和关键目录：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +tree --owner <owner> --repo <repo> --ref <branch> --format json
|
||||
gitlink-cli repo +readme --owner <owner> --repo <repo> --ref <branch> --format json
|
||||
```
|
||||
|
||||
如果 `repo +tree` 不可用，回退：
|
||||
|
||||
```bash
|
||||
gitlink-cli api GET /<owner>/<repo>/sub_entries --query 'filepath=&ref=<branch>' --format json
|
||||
```
|
||||
|
||||
3. 采集协作和发布信息：
|
||||
|
||||
```bash
|
||||
gitlink-cli issue +list --owner <owner> --repo <repo> --state open --format json
|
||||
gitlink-cli pr +list --owner <owner> --repo <repo> --state merged --format json
|
||||
gitlink-cli release +list --owner <owner> --repo <repo> --format json
|
||||
```
|
||||
|
||||
4. 按 `references/handbook-template.md` 生成科研成果沉淀手册。
|
||||
|
||||
## 输出格式
|
||||
|
||||
```markdown
|
||||
# 科研成果沉淀手册：<owner>/<repo>
|
||||
|
||||
## 1. 项目概览
|
||||
## 2. 研究问题与成果定位
|
||||
## 3. 代码与目录导览
|
||||
## 4. 实验和数据入口
|
||||
## 5. 复现步骤
|
||||
## 6. 协作与贡献记录
|
||||
## 7. 发布与引用信息
|
||||
## 8. 交接清单
|
||||
```
|
||||
|
||||
## 分析原则
|
||||
|
||||
- 手册是“可交接材料”，不是营销稿；优先列入口、证据和缺失项。
|
||||
- 不要编造论文标题、数据来源或实验结果；缺失时写“待补充”。
|
||||
- 对 Issue/PR 只做摘要，不贴完整讨论。
|
||||
- 如果仓库没有 Release，建议使用 Release 固化阶段成果。
|
||||
- 如果根目录 README 已完整，保留其结构并补交接清单；如果 README 很弱，按模板重建手册。
|
||||
|
||||
|
|
@ -0,0 +1,25 @@
|
|||
# Artifact handbook example
|
||||
|
||||
User request:
|
||||
|
||||
```text
|
||||
请为 Gitconomy/Git4Research 生成一份科研成果沉淀手册，用于课题组交接和答辩展示。
|
||||
```
|
||||
|
||||
Agent steps:
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner Gitconomy --repo Git4Research --format json
|
||||
gitlink-cli repo +readme --owner Gitconomy --repo Git4Research --ref main --format json
|
||||
gitlink-cli repo +tree --owner Gitconomy --repo Git4Research --ref main --format json
|
||||
gitlink-cli issue +list --owner Gitconomy --repo Git4Research --state open --format json
|
||||
gitlink-cli pr +list --owner Gitconomy --repo Git4Research --state merged --format json
|
||||
gitlink-cli release +list --owner Gitconomy --repo Git4Research --format json
|
||||
```
|
||||
|
||||
Expected answer:
|
||||
|
||||
- Produce a handbook using the template.
|
||||
- Mark missing evidence as `待补充`.
|
||||
- Confirm that no Wiki, Issue, PR, Release, or file write operation was performed.
|
||||
|
||||
|
|
@ -0,0 +1,59 @@
|
|||
# Research artifact handbook template
|
||||
|
||||
```markdown
|
||||
# 科研成果沉淀手册：<owner>/<repo>
|
||||
|
||||
## 1. 项目概览
|
||||
|
||||
- 仓库：
|
||||
- 默认分支：
|
||||
- 主要语言：
|
||||
- 维护状态：
|
||||
- 数据来源：
|
||||
|
||||
## 2. 研究问题与成果定位
|
||||
|
||||
- 研究问题：
|
||||
- 方法或系统：
|
||||
- 对应论文/报告：
|
||||
- 创新点：
|
||||
|
||||
## 3. 代码与目录导览
|
||||
|
||||
| 路径 | 作用 | 证据来源 |
|
||||
|---|---|---|
|
||||
|
||||
## 4. 实验和数据入口
|
||||
|
||||
| 内容 | 路径/链接 | 状态 | 待补充 |
|
||||
|---|---|---|---|
|
||||
|
||||
## 5. 复现步骤
|
||||
|
||||
1. 环境准备：
|
||||
2. 数据准备：
|
||||
3. 训练/运行：
|
||||
4. 评估：
|
||||
|
||||
## 6. 协作与贡献记录
|
||||
|
||||
- 贡献者：
|
||||
- 关键 Issue：
|
||||
- 已合并 PR：
|
||||
|
||||
## 7. 发布与引用信息
|
||||
|
||||
- Release：
|
||||
- License：
|
||||
- Citation：
|
||||
|
||||
## 8. 交接清单
|
||||
|
||||
- [ ] README 完整
|
||||
- [ ] LICENSE 明确
|
||||
- [ ] 依赖文件存在
|
||||
- [ ] 数据来源明确
|
||||
- [ ] 实验命令可运行
|
||||
- [ ] Release 或归档版本可引用
|
||||
```
|
||||
|
||||
|
|
@ -0,0 +1,79 @@
|
|||
---
|
||||
name: gitlink-research-collaboration-map
|
||||
description: "科研协作画像与合作建议：分析 GitLink 科研仓库的贡献者、Issue、PR、语言和用户信息，识别核心维护者、潜在协作者、未响应研究问题和协作分工建议。用于课题组协作复盘、跨团队合作匹配和学生贡献画像。"
|
||||
---
|
||||
|
||||
# gitlink-research-collaboration-map
|
||||
|
||||
开始前必须先阅读 `../gitlink-shared/SKILL.md`，确认认证、权限、安全规则和 GitLink API 注意事项。
|
||||
|
||||
## 安全规则
|
||||
|
||||
- 只读执行，不自动指派 Issue、不邀请成员。
|
||||
- 所有 GitLink 操作必须使用 `gitlink-cli`。
|
||||
- 所有命令使用 `--format json`。
|
||||
- 不输出 Token、Cookie 或认证 Header。
|
||||
- 不输出私人联系方式或敏感个人信息。
|
||||
- 协作建议必须基于公开仓库数据，避免做人身评价。
|
||||
|
||||
## 工作流
|
||||
|
||||
1. 获取仓库、语言、贡献者：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner <owner> --repo <repo> --format json
|
||||
gitlink-cli repo +languages --owner <owner> --repo <repo> --format json
|
||||
gitlink-cli repo +contributors --owner <owner> --repo <repo> --format json
|
||||
```
|
||||
|
||||
如 shortcut 不可用，回退：
|
||||
|
||||
```bash
|
||||
gitlink-cli api GET /<owner>/<repo>/languages --format json
|
||||
gitlink-cli api GET /<owner>/<repo>/contributors --format json
|
||||
```
|
||||
|
||||
2. 获取 Issue 和 PR：
|
||||
|
||||
```bash
|
||||
gitlink-cli issue +list --owner <owner> --repo <repo> --state open --format json
|
||||
gitlink-cli issue +list --owner <owner> --repo <repo> --state closed --format json
|
||||
gitlink-cli pr +list --owner <owner> --repo <repo> --state open --format json
|
||||
gitlink-cli pr +list --owner <owner> --repo <repo> --state merged --format json
|
||||
```
|
||||
|
||||
3. 对高频作者或贡献者，按需获取公开用户信息：
|
||||
|
||||
```bash
|
||||
gitlink-cli user +info --login <login> --format json
|
||||
```
|
||||
|
||||
4. 按 `references/profile-fields.md` 生成协作画像。
|
||||
|
||||
## 输出格式
|
||||
|
||||
```markdown
|
||||
# 科研协作画像：<owner>/<repo>
|
||||
|
||||
## 协作结构
|
||||
|
||||
| 角色 | 候选人/账号 | 证据 | 建议 |
|
||||
|---|---|---|---|
|
||||
| 核心维护者 | ... | commits/issues/pr | ... |
|
||||
|
||||
## 未响应问题
|
||||
|
||||
| Issue/PR | 主题 | 等待动作 | 建议协作者 |
|
||||
|---|---|---|---|
|
||||
|
||||
## 合作建议
|
||||
|
||||
1. ...
|
||||
```
|
||||
|
||||
## 分析原则
|
||||
|
||||
- 贡献者画像使用“贡献事实”和“适合跟进的任务”，不要给人格标签。
|
||||
- Issue/PR 作者、assignee、reviewer 字段可能缺失，缺失时不要强推断。
|
||||
- 镜像仓库的 GitLink 协作数据可能很少，需要标注局限。
|
||||
- 合作建议应给出证据：语言栈、历史贡献、Issue/PR 主题。
|
||||
|
|
@ -0,0 +1,23 @@
|
|||
# Collaboration map example
|
||||
|
||||
User request:
|
||||
|
||||
```text
|
||||
请分析 Gitconomy/Git4Research 的科研协作结构，找出核心维护者和可跟进的问题方向。
|
||||
```
|
||||
|
||||
Agent steps:
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner Gitconomy --repo Git4Research --format json
|
||||
gitlink-cli repo +contributors --owner Gitconomy --repo Git4Research --format json
|
||||
gitlink-cli issue +list --owner Gitconomy --repo Git4Research --state open --format json
|
||||
gitlink-cli pr +list --owner Gitconomy --repo Git4Research --state open --format json
|
||||
```
|
||||
|
||||
Expected answer:
|
||||
|
||||
- Separate facts from recommendations.
|
||||
- Mention data limits if the repository has little Issue/PR activity.
|
||||
- Do not expose private contact information.
|
||||
|
||||
|
|
@ -0,0 +1,20 @@
|
|||
# Collaboration profile fields
|
||||
|
||||
| 字段 | 来源 | 用途 |
|
||||
|---|---|---|
|
||||
| contributor login/name | `repo +contributors` | 识别代码贡献者 |
|
||||
| commits/contributions | `repo +contributors` | 粗略衡量贡献量 |
|
||||
| issue author | `issue +list` | 识别问题提出者 |
|
||||
| issue assignee | `issue +list` | 识别责任分配 |
|
||||
| issue labels | `issue +list` | 判断研究主题或任务类型 |
|
||||
| PR author | `pr +list` | 识别代码贡献者和集成者 |
|
||||
| PR status | `pr +list` | 判断协作流转状态 |
|
||||
| language | `repo +languages` | 推断技术栈 |
|
||||
|
||||
Suggested role labels:
|
||||
|
||||
- 核心维护者：代码贡献和 Issue/PR 参与均较多。
|
||||
- 问题提出者：Issue 活跃但代码贡献少，适合需求澄清和复现实验。
|
||||
- 实验贡献者：PR 或实验相关 Issue 活跃。
|
||||
- 潜在协作者：与主要语言或研究主题匹配，但当前参与较少。
|
||||
|
||||
|
|
@ -0,0 +1,77 @@
|
|||
---
|
||||
name: gitlink-research-compliance
|
||||
description: "科研开源合规与敏感风险检查：审计 GitLink 科研仓库的许可证、SECURITY、CONTRIBUTING、依赖声明和敏感文件风险。用于开源发布前预检查、论文代码发布合规清单和整改建议。"
|
||||
---
|
||||
|
||||
# gitlink-research-compliance
|
||||
|
||||
开始前必须先阅读 `../gitlink-shared/SKILL.md`，确认认证、权限、安全规则和 GitLink API 注意事项。
|
||||
|
||||
## 安全规则
|
||||
|
||||
- 只读执行，不删除文件、不改 `.gitignore`、不关闭 Issue。
|
||||
- 所有 GitLink 操作必须使用 `gitlink-cli`。
|
||||
- 所有命令使用 `--format json`。
|
||||
- 不输出 Token、Cookie 或认证 Header。
|
||||
- 报告必须注明“工程合规预审，不构成法律意见”。
|
||||
|
||||
## 工作流
|
||||
|
||||
1. 采集仓库信息和文件树：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner <owner> --repo <repo> --format json
|
||||
gitlink-cli repo +tree --owner <owner> --repo <repo> --ref <branch> --format json
|
||||
```
|
||||
|
||||
2. 如果 `repo +tree` 不可用，回退：
|
||||
|
||||
```bash
|
||||
gitlink-cli api GET /<owner>/<repo>/sub_entries --query 'filepath=&ref=<branch>' --format json
|
||||
```
|
||||
|
||||
3. 对根目录和高风险目录做规则检查：
|
||||
|
||||
- 根目录：许可证、安全策略、贡献指南、依赖清单、`.gitignore`。
|
||||
- 配置目录：`.github`、`.gitlink`、`.vscode`、`config`。
|
||||
- 脚本目录：`scripts`、`deploy`、`docker`。
|
||||
|
||||
4. 按 `references/risk-rules.md` 输出风险分级。
|
||||
|
||||
## 输出格式
|
||||
|
||||
```markdown
|
||||
# 科研开源合规预审：<owner>/<repo>
|
||||
|
||||
## 总览
|
||||
|
||||
| 项目 | 结果 | 风险等级 | 证据 |
|
||||
|---|---|---|---|
|
||||
|
||||
## 高优先级风险
|
||||
|
||||
1. ...
|
||||
|
||||
## 整改建议
|
||||
|
||||
1. ...
|
||||
|
||||
## 说明
|
||||
|
||||
本报告是工程合规预审，不构成法律意见。
|
||||
```
|
||||
|
||||
## 检查项
|
||||
|
||||
- 许可证：存在 `LICENSE`，README 中最好也声明许可证。
|
||||
- 安全策略：存在 `SECURITY.md`。
|
||||
- 贡献指南：存在 `CONTRIBUTING.md`。
|
||||
- 依赖声明：存在语言对应依赖文件，便于后续许可证和漏洞扫描。
|
||||
- 敏感文件风险：文件名包含 `.env`、`id_rsa`、`.pem`、`.key`、`credential`、`secret`、`token` 时列为风险。
|
||||
|
||||
## 注意事项
|
||||
|
||||
- 不要下载或展示疑似密钥内容，只报告路径和风险类型。
|
||||
- 如果只能获取根目录文件树，报告必须说明深层敏感文件未覆盖。
|
||||
- 私有仓库需要用户自行完成认证，不能要求用户粘贴 Token。
|
||||
|
||||
|
|
@ -0,0 +1,21 @@
|
|||
# Compliance audit example
|
||||
|
||||
User request:
|
||||
|
||||
```text
|
||||
请检查 Gitconomy/Git4Research 是否具备开源发布合规基础。
|
||||
```
|
||||
|
||||
Agent steps:
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner Gitconomy --repo Git4Research --format json
|
||||
gitlink-cli repo +tree --owner Gitconomy --repo Git4Research --ref main --format json
|
||||
```
|
||||
|
||||
Expected answer:
|
||||
|
||||
- Report license, security policy, contribution guide, dependency manifest, and sensitive file risks.
|
||||
- Mention whether the audit is root-only or includes selected subdirectories.
|
||||
- Avoid quoting or exposing sensitive file contents.
|
||||
|
||||
|
|
@ -0,0 +1,11 @@
|
|||
# Compliance risk rules
|
||||
|
||||
| 风险 | 等级 | 判定 | 建议 |
|
||||
|---|---|---|---|
|
||||
| 缺少 LICENSE | 高 | 根目录无 LICENSE 文件 | 补充许可证并在 README 声明 |
|
||||
| 疑似私钥 | 高 | 文件名包含 `id_rsa`、`.pem`、`.key` | 移除文件、轮换密钥、加入 `.gitignore` |
|
||||
| 疑似环境密钥 | 高 | 文件名包含 `.env`、`credential`、`secret`、`token` | 移除真实配置，保留 `.env.example` |
|
||||
| 缺少依赖声明 | 中 | 无常见依赖清单 | 补充依赖文件，便于许可证和漏洞审计 |
|
||||
| 缺少 SECURITY | 中 | 无 `SECURITY.md` | 补充漏洞披露流程 |
|
||||
| 缺少 CONTRIBUTING | 低 | 无 `CONTRIBUTING.md` | 补充贡献流程和代码规范 |
|
||||
|
||||
|
|
@ -0,0 +1,79 @@
|
|||
---
|
||||
name: gitlink-research-data-provenance
|
||||
description: "科研数据来源与隐私审计：检查 GitLink 科研仓库中的数据集说明、下载来源、引用链、隐私风险、数据许可证和大文件痕迹。用于论文代码发布前的数据治理、复现实验数据说明和科研合规预审。"
|
||||
---
|
||||
|
||||
# gitlink-research-data-provenance
|
||||
|
||||
开始前必须先阅读 `../gitlink-shared/SKILL.md`，确认认证、权限、安全规则和 GitLink API 注意事项。
|
||||
|
||||
## 安全规则
|
||||
|
||||
- 只读执行，不下载大文件、不打开疑似敏感数据内容。
|
||||
- 所有 GitLink 操作必须使用 `gitlink-cli`。
|
||||
- 所有命令使用 `--format json`。
|
||||
- 不输出 Token、Cookie 或认证 Header。
|
||||
- 不输出个人隐私、样本原文、密钥或数据内容；只报告路径、类型和风险。
|
||||
- 报告必须注明“科研数据治理预审，不构成法律意见或伦理审查结论”。
|
||||
|
||||
## 工作流
|
||||
|
||||
1. 采集仓库概览和默认分支：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner <owner> --repo <repo> --format json
|
||||
```
|
||||
|
||||
2. 采集根目录文件树：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +tree --owner <owner> --repo <repo> --ref <branch> --format json
|
||||
```
|
||||
|
||||
如果 `repo +tree` 不可用，回退：
|
||||
|
||||
```bash
|
||||
gitlink-cli api GET /<owner>/<repo>/sub_entries --query 'filepath=&ref=<branch>' --format json
|
||||
```
|
||||
|
||||
3. 如果根目录存在 `data`、`datasets`、`benchmark`、`experiments`、`notebooks`、`docs`、`README`、`paper`、`assets`，只采集一层目录结构：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +tree --owner <owner> --repo <repo> --path data --ref <branch> --format json
|
||||
gitlink-cli repo +tree --owner <owner> --repo <repo> --path docs --ref <branch> --format json
|
||||
```
|
||||
|
||||
4. 按 `references/provenance-rules.md` 生成数据来源和隐私风险报告。
|
||||
|
||||
## 输出格式
|
||||
|
||||
```markdown
|
||||
# 科研数据来源与隐私审计：<owner>/<repo>
|
||||
|
||||
## 数据治理总览
|
||||
|
||||
| 检查项 | 结果 | 风险 | 证据 |
|
||||
|---|---|---|---|
|
||||
|
||||
## 数据来源链
|
||||
|
||||
| 数据/目录 | 来源说明 | 许可证/引用 | 复现状态 |
|
||||
|---|---|---|---|
|
||||
|
||||
## 风险与整改
|
||||
|
||||
1. ...
|
||||
|
||||
## 说明
|
||||
|
||||
本报告是科研数据治理预审，不构成法律意见或伦理审查结论。
|
||||
```
|
||||
|
||||
## 分析原则
|
||||
|
||||
- 有数据目录但无来源说明时，风险高于“未发现数据目录”。
|
||||
- 发现 `.csv`、`.jsonl`、`.parquet`、`.h5`、`.npy`、`.zip`、`.tar` 等文件名时，只报告文件名和路径，不打印内容。
|
||||
- 发现 `patient`、`student`、`email`、`phone`、`idcard`、`address`、`face`、`medical` 等词时，标注隐私复核风险。
|
||||
- 如果仓库只包含数据下载脚本，检查脚本是否说明下载地址、数据许可证和引用方式。
|
||||
- 无法确认的事项标注“未确认”，不要编造数据来源。
|
||||
|
||||
|
|
@ -0,0 +1,22 @@
|
|||
# Data provenance audit example
|
||||
|
||||
User request:
|
||||
|
||||
```text
|
||||
请检查 songhui18/ICCV2021 是否有科研数据来源、数据许可证或隐私风险问题。
|
||||
```
|
||||
|
||||
Agent steps:
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner songhui18 --repo ICCV2021 --format json
|
||||
gitlink-cli repo +tree --owner songhui18 --repo ICCV2021 --ref master --format json
|
||||
gitlink-cli repo +tree --owner songhui18 --repo ICCV2021 --path data --ref master --format json
|
||||
```
|
||||
|
||||
Expected answer:
|
||||
|
||||
- Report only paths and risk types, not raw data contents.
|
||||
- Separate data source, license, citation, privacy, and large-file risks.
|
||||
- State whether the assessment is root-only or includes selected subdirectories.
|
||||
|
||||
|
|
@ -0,0 +1,12 @@
|
|||
# Data provenance rules
|
||||
|
||||
| 检查项 | 风险等级 | 判定线索 | 建议 |
|
||||
|---|---|---|---|
|
||||
| 数据来源说明缺失 | 高 | 存在 data/datasets/benchmark 目录，但 README 或目录下无来源说明 | 补充数据集名称、下载地址、版本、引用论文和访问日期 |
|
||||
| 数据许可证缺失 | 高 | 存在数据文件或下载脚本，但无 license/citation/terms | 补充数据许可证、使用条款和引用格式 |
|
||||
| 隐私数据线索 | 高 | 文件名或目录包含 patient、student、email、phone、idcard、address、face、medical 等 | 进行脱敏和伦理审查，避免公开原始个人数据 |
|
||||
| 大文件或压缩包 | 中 | 文件名包含 .zip、.tar、.h5、.npy、.parquet 等 | 说明文件来源、校验和、下载方式，不建议直接托管大数据 |
|
||||
| 数据下载脚本无校验 | 中 | 有 download/get_data 脚本但无 checksum/version | 补充 checksum、版本号和失败处理 |
|
||||
| 引用链不完整 | 中 | README 未说明数据、论文、代码之间的关系 | 补充数据引用、论文引用和实验脚本映射 |
|
||||
| 未发现数据痕迹 | 低 | 未见数据目录、下载脚本或数据扩展名 | 在报告中标注未发现，建议说明数据是否需另行申请 |
|
||||
|
||||
|
|
@ -0,0 +1,90 @@
|
|||
---
|
||||
name: gitlink-research-knowledge-graph
|
||||
description: "科研热点追踪与轻量知识图谱：按关键词搜索 GitLink 科研仓库，聚合仓库、语言、贡献者、Issue/PR 主题和研究方向，输出趋势分析、图谱节点边和选题建议。用于科研选题、技术生态调研和领域知识图谱构建。"
|
||||
---
|
||||
|
||||
# gitlink-research-knowledge-graph
|
||||
|
||||
开始前必须先阅读 `../gitlink-shared/SKILL.md`，确认认证、权限、安全规则和 GitLink API 注意事项。
|
||||
|
||||
## 安全规则
|
||||
|
||||
- 只读执行。
|
||||
- 所有 GitLink 操作必须使用 `gitlink-cli`。
|
||||
- 所有命令使用 `--format json`。
|
||||
- 不输出 Token、Cookie 或认证 Header。
|
||||
- 控制 API 调用量：每个关键词保留前 8 个仓库，深度分析不超过 8 个仓库。
|
||||
- 对镜像仓库标注数据局限，不把 GitLink 平台内 watcher/fork 等同于原平台全量影响力。
|
||||
|
||||
## 工作流
|
||||
|
||||
1. 将研究主题拆成 3 到 5 个关键词，覆盖中文、英文、缩写和技术术语。
|
||||
|
||||
2. 搜索仓库：
|
||||
|
||||
```bash
|
||||
gitlink-cli search +repos -k <keyword> --format json
|
||||
```
|
||||
|
||||
3. 去重并选取候选仓库：
|
||||
|
||||
- 唯一键优先使用 `author.login/identifier`。
|
||||
- 同一仓库匹配多个关键词时保留并记录全部关键词。
|
||||
- 总候选控制在 20 个以内。
|
||||
- 深度分析优先选择匹配关键词多、更新时间近、关注度高的 5 到 8 个。
|
||||
|
||||
4. 对重点仓库采集详情：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner <owner> --repo <repo> --format json
|
||||
gitlink-cli repo +languages --owner <owner> --repo <repo> --format json
|
||||
gitlink-cli repo +contributors --owner <owner> --repo <repo> --format json
|
||||
gitlink-cli issue +list --owner <owner> --repo <repo> --state open --format json
|
||||
gitlink-cli pr +list --owner <owner> --repo <repo> --state open --format json
|
||||
```
|
||||
|
||||
5. 按 `references/graph-schema.md` 输出图谱 JSON 和趋势报告。
|
||||
|
||||
## 输出格式
|
||||
|
||||
```markdown
|
||||
# 科研热点与知识图谱：<topic>
|
||||
|
||||
## 搜索范围
|
||||
|
||||
关键词：...
|
||||
命中仓库：...
|
||||
深度分析：...
|
||||
|
||||
## 趋势洞察
|
||||
|
||||
1. ...
|
||||
|
||||
## 图谱摘要
|
||||
|
||||
| 节点类型 | 数量 |
|
||||
|---|---:|
|
||||
| repo | ... |
|
||||
| language | ... |
|
||||
| contributor | ... |
|
||||
| topic | ... |
|
||||
|
||||
## 选题建议
|
||||
|
||||
1. ...
|
||||
```
|
||||
|
||||
同时输出或附带如下 JSON 结构：
|
||||
|
||||
```json
|
||||
{
|
||||
"nodes": [],
|
||||
"edges": []
|
||||
}
|
||||
```
|
||||
|
||||
## 分析原则
|
||||
|
||||
- 不要把搜索结果数量直接解释为学术热度，只能作为 GitLink 平台内信号。
|
||||
- 图谱边必须有来源命令或字段依据。
|
||||
- 选题建议要区分“成熟方向”“新兴方向”“生态空白”。
|
||||
|
|
@ -0,0 +1,24 @@
|
|||
# Keyword knowledge graph example
|
||||
|
||||
User request:
|
||||
|
||||
```text
|
||||
请围绕“论文复现”和“open research”在 GitLink 上做科研热点搜索，并输出轻量知识图谱。
|
||||
```
|
||||
|
||||
Agent steps:
|
||||
|
||||
```bash
|
||||
gitlink-cli search +repos -k "论文复现" --format json
|
||||
gitlink-cli search +repos -k "open research" --format json
|
||||
gitlink-cli repo +info --owner <owner> --repo <repo> --format json
|
||||
gitlink-cli repo +languages --owner <owner> --repo <repo> --format json
|
||||
gitlink-cli repo +contributors --owner <owner> --repo <repo> --format json
|
||||
```
|
||||
|
||||
Expected answer:
|
||||
|
||||
- List keywords and candidate repositories.
|
||||
- Provide a Markdown trend report.
|
||||
- Provide `nodes` and `edges` JSON using `references/graph-schema.md`.
|
||||
|
||||
|
|
@ -0,0 +1,37 @@
|
|||
# Research knowledge graph schema
|
||||
|
||||
Node types:
|
||||
|
||||
| Type | Required fields | Description |
|
||||
|---|---|---|
|
||||
| `repo` | `id`, `owner`, `repo`, `label` | GitLink repository |
|
||||
| `language` | `id`, `name` | Programming language |
|
||||
| `contributor` | `id`, `login` | Public contributor identity |
|
||||
| `topic` | `id`, `name` | Keyword or inferred research topic |
|
||||
| `issue` | `id`, `title`, `state` | Representative Issue |
|
||||
| `pr` | `id`, `title`, `state` | Representative PR |
|
||||
|
||||
Edge types:
|
||||
|
||||
| Type | From | To | Evidence |
|
||||
|---|---|---|---|
|
||||
| `uses_language` | repo | language | `repo +languages` |
|
||||
| `contributed_by` | contributor | repo | `repo +contributors` |
|
||||
| `matches_topic` | repo | topic | search keyword or description |
|
||||
| `has_issue` | repo | issue | `issue +list` |
|
||||
| `has_pr` | repo | pr | `pr +list` |
|
||||
|
||||
Example:
|
||||
|
||||
```json
|
||||
{
|
||||
"nodes": [
|
||||
{"id": "repo:Gitconomy/Git4Research", "type": "repo", "label": "Gitconomy/Git4Research"},
|
||||
{"id": "topic:open-research", "type": "topic", "label": "open research"}
|
||||
],
|
||||
"edges": [
|
||||
{"source": "repo:Gitconomy/Git4Research", "target": "topic:open-research", "type": "matches_topic", "evidence": "search +repos keyword"}
|
||||
]
|
||||
}
|
||||
```
|
||||
|
||||
|
|
@ -0,0 +1,82 @@
|
|||
---
|
||||
name: gitlink-research-progress-tracker
|
||||
description: "科研项目进度跟踪与预警：基于 GitLink Issue、PR、Release、贡献者和仓库更新时间生成课题组周报、停滞风险、开放问题队列和下一步建议。用于科研项目管理、中期检查和例会汇报。"
|
||||
---
|
||||
|
||||
# gitlink-research-progress-tracker
|
||||
|
||||
开始前必须先阅读 `../gitlink-shared/SKILL.md`，确认认证、权限、安全规则和 GitLink API 注意事项。
|
||||
|
||||
## 安全规则
|
||||
|
||||
- 默认只读执行。
|
||||
- 所有 GitLink 操作必须使用 `gitlink-cli`。
|
||||
- 所有命令使用 `--format json`。
|
||||
- 不输出 Token、Cookie 或认证 Header。
|
||||
- 不自动关闭 Issue、不合并 PR、不发布周报评论。
|
||||
- 如果用户要求发布周报到 Issue，先生成 dry-run 草稿并等待确认。
|
||||
|
||||
## 工作流
|
||||
|
||||
1. 获取仓库概览：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner <owner> --repo <repo> --format json
|
||||
```
|
||||
|
||||
2. 获取协作队列：
|
||||
|
||||
```bash
|
||||
gitlink-cli issue +list --owner <owner> --repo <repo> --state open --format json
|
||||
gitlink-cli issue +list --owner <owner> --repo <repo> --state closed --format json
|
||||
gitlink-cli pr +list --owner <owner> --repo <repo> --state open --format json
|
||||
gitlink-cli pr +list --owner <owner> --repo <repo> --state merged --format json
|
||||
```
|
||||
|
||||
3. 获取版本和贡献者信息：
|
||||
|
||||
```bash
|
||||
gitlink-cli release +list --owner <owner> --repo <repo> --format json
|
||||
gitlink-cli repo +contributors --owner <owner> --repo <repo> --format json
|
||||
```
|
||||
|
||||
如果当前 CLI 版本没有 `repo +contributors`，回退：
|
||||
|
||||
```bash
|
||||
gitlink-cli api GET /<owner>/<repo>/contributors --format json
|
||||
```
|
||||
|
||||
4. 按 `references/risk-model.md` 判断风险，并生成周报。
|
||||
|
||||
## 输出格式
|
||||
|
||||
```markdown
|
||||
# 科研项目周报：<owner>/<repo>
|
||||
|
||||
## 本期概览
|
||||
|
||||
| 指标 | 数值 |
|
||||
|---|---:|
|
||||
| 开放 Issue | ... |
|
||||
| 已关闭 Issue | ... |
|
||||
| 开放 PR | ... |
|
||||
| 已合并 PR | ... |
|
||||
| Release 数 | ... |
|
||||
| 贡献者 | ... |
|
||||
|
||||
## 风险预警
|
||||
|
||||
| 风险 | 等级 | 证据 | 建议 |
|
||||
|---|---|---|---|
|
||||
|
||||
## 下一步建议
|
||||
|
||||
1. ...
|
||||
```
|
||||
|
||||
## 分析原则
|
||||
|
||||
- 先报告事实，再给建议。
|
||||
- 把“没有数据”和“数据为 0”区分开。
|
||||
- 如果 PR/Issue 的 state 过滤不精确，根据返回字段二次归类，并在报告中说明。
|
||||
- 重点识别阻塞项：长期开放 Issue、开放 PR 堆积、无 Release、近期无维护活动、单人维护风险。
|
||||
|
|
@ -0,0 +1,25 @@
|
|||
# Research weekly report example
|
||||
|
||||
User request:
|
||||
|
||||
```text
|
||||
请给 songhui18/ICCV2021 生成一份科研项目进度周报和风险预警。
|
||||
```
|
||||
|
||||
Agent steps:
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner songhui18 --repo ICCV2021 --format json
|
||||
gitlink-cli issue +list --owner songhui18 --repo ICCV2021 --state open --format json
|
||||
gitlink-cli issue +list --owner songhui18 --repo ICCV2021 --state closed --format json
|
||||
gitlink-cli pr +list --owner songhui18 --repo ICCV2021 --state open --format json
|
||||
gitlink-cli pr +list --owner songhui18 --repo ICCV2021 --state merged --format json
|
||||
gitlink-cli release +list --owner songhui18 --repo ICCV2021 --format json
|
||||
```
|
||||
|
||||
Expected answer:
|
||||
|
||||
- Provide a concise weekly report.
|
||||
- Identify the top 3 risks with evidence.
|
||||
- Do not write comments back unless explicitly requested.
|
||||
|
||||
|
|
@ -0,0 +1,11 @@
|
|||
# Research progress risk model
|
||||
|
||||
| 风险 | 等级 | 判定线索 | 建议 |
|
||||
|---|---|---|---|
|
||||
| Issue 堆积 | 中/高 | 开放 Issue 明显多于关闭 Issue，或关键问题长期未响应 | 分类 Issue，标注优先级，安排负责人 |
|
||||
| PR 堆积 | 中/高 | 开放 PR 多且缺少 review/merge 记录 | 建立 review 节奏，先处理小 PR |
|
||||
| 版本停滞 | 中 | 无 Release 或长期无 Release | 为阶段性实验结果建立 Release |
|
||||
| 维护停滞 | 高 | 仓库更新时间较久且仍有开放问题 | 明确是否归档、继续维护或招募维护者 |
|
||||
| 单点维护 | 中 | 贡献者数量很少 | 补贡献指南和 onboarding 任务 |
|
||||
| 数据不足 | 低 | API 缺失或仓库未启用对应功能 | 标注数据不可用，不推断过度结论 |
|
||||
|
||||
|
|
@ -0,0 +1,90 @@
|
|||
---
|
||||
name: gitlink-research-reproducibility
|
||||
description: "科研仓库复现性审计：检查 GitLink 科研、论文复现、实验代码仓库是否具备 README、LICENSE、依赖清单、测试入口、CI、示例和数据/实验说明。用于生成复现性评分、缺失项和整改建议。"
|
||||
---
|
||||
|
||||
# gitlink-research-reproducibility
|
||||
|
||||
开始前必须先阅读 `../gitlink-shared/SKILL.md`，确认认证、权限、安全规则和 GitLink API 注意事项。
|
||||
|
||||
## 安全规则
|
||||
|
||||
- 只读执行，不创建 Issue、不评论、不修改仓库。
|
||||
- 所有 GitLink 操作必须使用 `gitlink-cli`。
|
||||
- 所有命令使用 `--format json`。
|
||||
- 不输出 Token、Cookie 或认证 Header。
|
||||
- 如果用户要求回写整改清单，先输出 dry-run 文本并等待确认。
|
||||
|
||||
## 工作流
|
||||
|
||||
1. 确认目标仓库：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner <owner> --repo <repo> --format json
|
||||
```
|
||||
|
||||
2. 识别默认分支。优先使用 `repo +info` 返回的 `default_branch` 或 `default_branch_name`。没有字段时再询问用户，或使用 `master` 作为 GitLink 常见默认值。
|
||||
|
||||
3. 采集根目录文件树：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +tree --owner <owner> --repo <repo> --ref <branch> --format json
|
||||
```
|
||||
|
||||
如果当前 CLI 版本没有 `repo +tree`，回退 Raw API：
|
||||
|
||||
```bash
|
||||
gitlink-cli api GET /<owner>/<repo>/sub_entries --query 'filepath=&ref=<branch>' --format json
|
||||
```
|
||||
|
||||
4. 如根目录存在 `docs`、`examples`、`data`、`experiments`、`notebooks`、`tests`、`.github`、`.gitlink`，按需采集一层子目录用于证据补强：
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +tree --owner <owner> --repo <repo> --path docs --ref <branch> --format json
|
||||
```
|
||||
|
||||
5. 按 `references/checklist.md` 的评分表生成报告。没有读取子目录证据时，报告必须标注“仅基于根目录文件树”。
|
||||
|
||||
## 输出格式
|
||||
|
||||
```markdown
|
||||
# 科研复现性审计：<owner>/<repo>
|
||||
|
||||
## 总分
|
||||
|
||||
<score>/100
|
||||
|
||||
## 检查明细
|
||||
|
||||
| 项目 | 分值 | 结果 | 证据 | 建议 |
|
||||
|---|---:|---|---|---|
|
||||
| README | 20 | 通过/缺失 | root_entries | ... |
|
||||
|
||||
## 优先整改
|
||||
|
||||
1. ...
|
||||
2. ...
|
||||
3. ...
|
||||
|
||||
## 数据来源
|
||||
|
||||
- `repo +info`
|
||||
- `repo +tree` 或 `api GET /sub_entries`
|
||||
```
|
||||
|
||||
## 判定规则
|
||||
|
||||
- README：`README.md`、`README.zh-CN.md`、`readme`。
|
||||
- LICENSE：`LICENSE`、`LICENSE.md`、`LICENSE.txt`。
|
||||
- 依赖清单：`requirements.txt`、`pyproject.toml`、`environment.yml`、`package.json`、`go.mod`、`Cargo.toml`、`pom.xml` 等。
|
||||
- 测试入口：`tests`、`test`、`__tests__`。
|
||||
- CI 配置：`.github`、`.gitlink`、`.gitlab-ci.yml`、`Jenkinsfile`、`.circleci`。
|
||||
- 示例或文档：`docs`、`examples`、`demo`。
|
||||
- 数据或实验说明：`data`、`datasets`、`experiments`、`notebooks`。
|
||||
|
||||
## 注意事项
|
||||
|
||||
- 复现性评分是工程辅助判断，不等同于论文质量评价。
|
||||
- GitLink 平台 API 可能返回字段差异，缺失字段时用“未知”而不是编造。
|
||||
- 如果仓库是镜像仓库，需标注 GitLink 内协作数据可能不代表原平台活跃度。
|
||||
|
||||
|
|
@ -0,0 +1,23 @@
|
|||
# Reproducibility audit example
|
||||
|
||||
User request:
|
||||
|
||||
```text
|
||||
请用 GitLink 分析 songhui18/ICCV2021 的论文复现性，输出缺失项和整改优先级。
|
||||
```
|
||||
|
||||
Agent steps:
|
||||
|
||||
```bash
|
||||
gitlink-cli repo +info --owner songhui18 --repo ICCV2021 --format json
|
||||
gitlink-cli repo +tree --owner songhui18 --repo ICCV2021 --ref master --format json
|
||||
gitlink-cli api GET /songhui18/ICCV2021/sub_entries --query 'filepath=&ref=master' --format json
|
||||
```
|
||||
|
||||
Expected answer:
|
||||
|
||||
- Quote the collected command names.
|
||||
- Score README, LICENSE, dependencies, tests, CI, examples/docs, data notes.
|
||||
- List the top three fixes before any optional polishing.
|
||||
- State that no write action was performed.
|
||||
|
||||
|
|
@ -0,0 +1,12 @@
|
|||
# Reproducibility checklist
|
||||
|
||||
| 项目 | 分值 | 通过条件 | 缺失时建议 |
|
||||
|---|---:|---|---|
|
||||
| README | 20 | 根目录存在 README，且能说明项目目标、安装或运行方式 | 补充项目背景、环境准备、最小运行命令 |
|
||||
| LICENSE | 15 | 根目录存在 LICENSE 文件 | 选择合适开源许可证并在 README 中声明 |
|
||||
| 依赖清单 | 15 | 存在语言对应依赖文件 | 补充 requirements.txt、pyproject.toml、environment.yml、go.mod 等 |
|
||||
| 测试入口 | 20 | 存在 tests/test/__tests__ 或可运行验证脚本 | 提供最小单元测试或 smoke test |
|
||||
| CI 配置 | 10 | 存在 CI 配置目录或文件 | 增加基础 lint/test CI |
|
||||
| 示例或文档 | 10 | 存在 docs/examples/demo | 补充复现实验步骤或示例输入输出 |
|
||||
| 数据或实验说明 | 10 | 存在 data/datasets/experiments/notebooks 或 README 中明确说明 | 写明数据来源、下载方式、实验配置 |
|
||||
|
||||
Loading…
Reference in New Issue