PR Babysitter
openinterpreter/openinterpreter
Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.
Palantir本体论实践版本 v3.5 - 强制执行版。结构化知识图谱+规则引擎+科学推理方法论。每次决策/分析前必须使用。推理结果必须展示详细推理过程:包含用户需求、规则依据ID、规则内容、置信度标注。
$ npx skills add LeoYeAI/openclaw-master-skills --skill ontology-clawra -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install LeoYeAI/openclaw-master-skills ontology-clawra --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/ontology-clawra-backup-20260319-151919 .claude/skills/ontology-clawra && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "ontology-clawra" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/ontology-clawra-backup-20260319-151919 into .claude/skills/ontology-clawra/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ontology-clawra", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/ontology-clawra-backup-20260319-151919Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add LeoYeAI/openclaw-master-skills --skill ontology-clawra -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install LeoYeAI/openclaw-master-skills ontology-clawra --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/ontology-clawra-backup-20260319-151919 .agents/skills/ontology-clawra && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "ontology-clawra" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/ontology-clawra-backup-20260319-151919 into .agents/skills/ontology-clawra/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ontology-clawra", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add LeoYeAI/openclaw-master-skills --skill ontology-clawra -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install LeoYeAI/openclaw-master-skills ontology-clawra --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/ontology-clawra-backup-20260319-151919 .cursor/skills/ontology-clawra && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "ontology-clawra" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/ontology-clawra-backup-20260319-151919 into .cursor/skills/ontology-clawra/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ontology-clawra", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/LeoYeAI/openclaw-master-skills.git --path skills/ontology-clawra-backup-20260319-151919--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add LeoYeAI/openclaw-master-skills --skill ontology-clawra -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install LeoYeAI/openclaw-master-skills ontology-clawra --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/ontology-clawra-backup-20260319-151919 .gemini/skills/ontology-clawra && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "ontology-clawra" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/ontology-clawra-backup-20260319-151919 into .gemini/skills/ontology-clawra/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ontology-clawra", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install LeoYeAI/openclaw-master-skills ontology-clawraInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add LeoYeAI/openclaw-master-skills --skill ontology-clawra -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/ontology-clawra-backup-20260319-151919 .github/skills/ontology-clawra && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "ontology-clawra" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/ontology-clawra-backup-20260319-151919 into .github/skills/ontology-clawra/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ontology-clawra", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add LeoYeAI/openclaw-master-skills --skill ontology-clawra -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install LeoYeAI/openclaw-master-skills ontology-clawra --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/ontology-clawra-backup-20260319-151919 .opencode/skills/ontology-clawra && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "ontology-clawra" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/ontology-clawra-backup-20260319-151919 into .opencode/skills/ontology-clawra/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ontology-clawra", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
ontology-clawraPalantir本体论实践版本 v3.5 - 强制执行版。结构化知识图谱+规则引擎+科学推理方法论。每次决策/分析前必须使用。推理结果必须展示详细推理过程:包含用户需求、规则依据ID、规则内容、置信度标注。
Ontology Clawra is an agent skill from LeoYeAI/openclaw-master-skills. Palantir本体论实践版本 v3.5 - 强制执行版。结构化知识图谱+规则引擎+科学推理方法论。每次决策/分析前必须使用。推理结果必须展示详细推理过程:包含用户需求、规则依据ID、规则内容、置信度标注。 强制执行原则(v3.5 新增): - 所有推理任务,必须首先使用本技能框架,禁止凭经验输出 - 本技能用于:推理过程记录、本体完善、规则优化 - 本技能不包含用户私人数据,数据严格本地私有 - 本技能与 ontology-platform 项目严格分离:技能是内核,项目是衍生 安全边界(重要): - 自动学习:默认禁用,需用户明确授权"写入本体" - 文件读取:仅读取 ~/.openclaw/skills/ontology-clawra/memory/ - 文件写入:仅在用户确认后写入本地本体 - 网络:不主动发起网络请求 - 隐私:绝不上传用户数据到任何服务器 - 红线:用户私人推理记录、个人决策数据严禁同步到GitHub等外部平台
Its SKILL.md is about 4.7k tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files (for example `_meta.json`, `laws.yaml` and `laws_detector.yaml`).
It works with GitHub. The repository describes itself as: 🧠 Curated collection of 1209+ best OpenClaw skills — weekly updated by MyClaw.ai. The licence is MIT.
4 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit e5199b5. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Shell commands in SKILL.md call:
python3From the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Ontology Clawra loads about 4.7k tokens when it runs. Until then it costs about 120 tokens; SKILL.md has 287 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from LeoYeAI/openclaw-master-skills at commit e5199b5, republished under its MIT licence (© LeoYeAI). 287 words, ~4,669 tokens.
.claude/skills/ontology-clawra/SKILL.md (or your agent's skills folder). This skill also uses 5 other files; get the full folder from GitHub.Palantir本体论实践版本 - Clawra的核心智能引擎 v3.5
| 功能 | 默认状态 | 触发条件 |
|---|---|---|
| 自动抽取到本体 | 🔴 禁用 | 用户明确确认"写入本体" |
| 自动置信度升级 | 🔴 禁用 | 用户确认推理结果正确 |
| 读取工作区文件 | 🟡 受限 | 仅读取memory/目录 |
| 写入本地本体 | 🔴 禁用 | 用户确认后单次执行 |
memory/ 目录,不读取其他敏感文件| v3.2 | v3.3 (新增) |
|---|---|
| 用户不说"抽取" → 不提取 | 用户确认推理 → 自动提取 |
| 重复概念 → 忽略 | 重复出现3次 → 自动识别 |
| 推理失败 → 等待用户问 | 推理失败 → 主动建议补充 |
AUTO_LEARN_TRIGGERS:
# 触发1:用户明确说"写入本体"或"记录这个"
- event: "user_says_write_ontology"
action: "extract_to_ontology"
requires_confirmation: true # 每次写入前必须确认
# 触发2:用户说"确认这个是对的"
- event: "user_confirms_reasoning"
action: "suggest_upgrade_confidence"
requires_confirmation: true # 建议升级,但需用户确认
# 触发3:推理失败(仅提示,不自动写入)
- event: "ontology_lookup_failed"
action: "suggest_supplement"
prompt_user: true
auto_write: false # 不自动写入
# 触发4:用户纠正错误(仅记录,不自动修改)
- event: "user_correction"
action: "log_correction"
auto_write: false # 不自动修改本体
### 自动抽取流程(⚠️ 每次写入需用户确认)
用户明确说"写入本体"或"记录这个" │ ▼ ┌─────────────────────────────┐ │ 1. 识别可抽取内容 │ │ - 新概念 (Concept) │ │ - 新规律 (Law) │ │ - 新规则 (Rule) │ └─────────────┬───────────────┘ │ ▼ ┌─────────────────────────────┐ │ 2. 展示给用户确认 │ ⚠️ 关键! │ "即将写入以下内容:xxx" │ └─────────────┬───────────────┘ │ 用户确认"是的" ▼ ┌─────────────────────────────┐ │ 3. 写入本体 │ └─────────────┬───────────────┘ │ ▼ ┌─────────────────────────────┐ │ 4. 反馈用户 │ │ "已写入本体:xxx" │ └─────────────────────────────┘
---
## 🧬 核心理念升级
### v2.0 问题
- 纯架构设计,缺少方法论
- 无本体自动构建能力
- 推理"照本宣科",缺乏科学性验证
### v3.0 改进
- ✅ 嵌入科学推理方法论
- ✅ 支持交互式本体构建
- ✅ 平衡灵活性与科学性
### v3.3 升级
- ✅ **主动学习**:用户确认后自动抽取
- ✅ **智能触发**:高频实体自动识别
- ✅ **推理失败建议**:主动提示补充本体
---
## 一、科学推理方法论(必读)
### ⚠️ 任何推理前必须遵循的流程
┌─────────────────────────────────────────────────────────────┐ │ 推理前置检查流程 │ ├─────────────────────────────────────────────────────────────┤ │ │ │ 1️⃣ 检查本体(Check) │ │ ↓ │ │ 本体有相关数据?→ 调用本体推理 │ │ ↓ 无 │ │ ↓ │ │ 2️⃣ 声明来源(Declare) │ │ "以下为外部知识/猜测/假设,需要验证" │ │ ↓ │ │ 3️⃣ 交互确认(Confirm)- ⚠️ 关键步骤! │ │ 关键假设必须用户确认后再深入 │ │ ↓ │ │ 4️⃣ 标注假设(Label) │ │ 明确标注哪些是"推测"、哪些是"确认" │ │ ↓ │ │ 5️⃣ 灵活推理(Reason) │ │ 结合本体 + 合理假设 + 明确标注 │ │ │ └─────────────────────────────────────────────────────────────┘
### ⚠️ 铁律:不确定时必须交互确认
当推理过程中存在以下情况时,必须暂停并与用户确认: ─────────────────────────────────────────────────────────── ❌ 禁止直接输出结论的情况:
存在 ASSUMED 级别的关键假设 → 必须问用户确认后才能给出最终结论
缺少必要的输入参数 → 必须先询问用户获取必要信息
计算结果依赖多个假设 → 必须列出所有假设,让用户确认
给出多个方案但无法确定最优 → 必须让用户选择或确认偏好
✅ 正确的交互流程:
Step 1: 列出已确认的信息(本体数据) Step 2: 列出不确定的信息(需要确认的假设) Step 3: 提供典型场景/默认值供选择 Step 4: 等待用户确认后再输出最终结论
⚡ 违规判定:
### 推理结果可信度标注
| 标注 | 含义 | 行动 |
|------|------|------|
| 🟢 **CONFIRMED** | 本体/记忆中有确凿数据 | 直接使用 |
| 🟡 **ASSUMED** | 基于合理假设,未验证 | 需用户确认 |
| 🔴 **SPECULATIVE** | 纯猜测,无依据 | 明确声明,谨慎使用 |
| ⚪ **UNKNOWN** | 确实不知道 | 坦诚告知用户 |
---
## 二、四大支柱(保留并增强)
### 2.1 Objects(对象)
```yaml
# 主体
Person:
- id, name, role, goals[], preferences{}, capabilities[]
# 概念/抽象
Concept:
- id, name, definition, examples[], properties{}
# 规律/法则
Law:
- id, name, domain, statement, conditions[], effects[], source, confidence
# 意图
Objective:
- id, name, priority, criteria{}, status
# 项目
Project:
- id, name, objectives[], status, owner, depends_on[]
# 任务
Task:
- id, title, status, priority, assignee, blockers[], estimated_hours
# 规则
Rule:
- id, name, condition, action, enabled, weight, source, confidence
# 决策
Decision:
- id, context, options[], selected, rationale, based_on_rules[], confidence# 基础关系
works_on: Person → Project/Task
depends_on: Task/Project → Task/Project
has_objective: Project → Objective
has_rule: Project/Objective → Rule
# 知识关系
exemplifies: Concept → Example
governs: Law → Domain
explains: Concept → Law
supports: Evidence → Rule
contradicts: Fact → Rule
derived_from: Rule/Law → Evidence # 新增:规则/规律的数据来源
# 推理关系
triggers: Rule → Decision
validates: Rule → Decision
refines: Rule → Rule
# 元关系
relates_to: Any → Any
is_a: Concept → Concept
part_of: Object → Object# 推理引擎核心函数(增强版)
def check_ontology(query):
"""1. 检查本体是否有相关数据"""
results = search_objects(query) + search_laws(query) + search_rules(query)
if results:
return {"status": "FOUND", "data": results, "confidence": "CONFIRMED"}
return {"status": "NOT_FOUND", "data": None, "confidence": "UNKNOWN"}
def declare_source(confidence_level, content):
"""2. 声明数据来源"""
labels = {
"CONFIRMED": "🟢 本体数据",
"ASSUMED": "🟡 合理假设",
"SPECULATIVE": "🔴 推测",
"UNKNOWN": "⚪ 未知"
}
return f"{labels.get(confidence_level, '')} {content}"
def confirm_with_user(assumptions):
"""3. 交互确认关键假设"""
# 返回需要确认的问题列表
return [f"请确认: {a}" for a in assumptions]
def label_result(content, confidence):
"""4. 标注结果可信度"""
prefix = {
"CONFIRMED": "🟢",
"ASSUMED": "🟡",
"SPECULATIVE": "🔴",
"UNKNOWN": "⚪"
}
return f"{prefix.get(confidence, '')} {content}"
def flexible_reasoning(ontology_data, assumptions, user_confirmations):
"""5. 灵活推理 - 结合本体+假设+确认"""
# 如果本体有数据,优先使用
# 如果需要假设,明确标注
# 如果用户已确认,升级置信度
pass
# 链式推理
def chain_reasoning(facts, rules, confidence_threshold=0.5):
"""链式推理:事实 + 规则 → 新结论"""
conclusions = []
for rule in rules:
if rule.confidence < confidence_threshold:
continue
if all(fact_matcher(f, rule.conditions) for f in facts):
conclusion = infer(rule, facts)
conclusion.source = f"derived_from:{rule.id}"
conclusion.confidence = min(rule.confidence, min(f.confidence for f in facts))
conclusions.append(conclusion)
return conclusions# 操作类型(增强)
Action:
- type: execute # 执行具体任务
- type: reason # 推理分析(走方法论流程)
- type: decide # 决策选择
- type: learn # 学习新知识(构建本体)
- type: validate # 验证一致性
- type: query # 查询知识网络
- type: extract # 新增:从交互中抽取本体
- type: confirm # 新增:请求用户确认┌─────────────────────────────────────────────────────────────┐
│ 本体自动构建流程(推荐使用) │
├─────────────────────────────────────────────────────────────┤
│ │
│ 用户输入 │
│ │ │
│ ▼ │
│ ┌─────────────────────┐ │
│ │ 实体识别 │ ← 识别可抽取的对象 │
│ │ - Person │ │
│ │ - Concept │ │
│ │ - Law │ │
│ │ - Rule │ │
│ └─────────┬───────────┘ │
│ │ │
│ ▼ │
│ ┌─────────────────────┐ │
│ │ 关系识别 │ ← 识别实体间关系 │
│ │ - is_a │ │
│ │ - relates_to │ │
│ │ - triggers │ │
│ │ - supports │ │
│ └─────────┬───────────┘ │
│ │ │
│ ▼ │
│ ┌─────────────────────┐ │
│ │ 去重检查 │ ← 避免重复构建 │
│ │ - 检查name是否已存在│ │
│ │ - 检查similar关系 │ │
│ └─────────┬───────────┘ │
│ │ │
│ ▼ │
│ ┌─────────────────────┐ │
│ │ 写入本体 │ ← 增量更新 │
│ │ - Objects │ │
│ │ - Links │ │
│ │ - 记录来源 │ │
│ └─────────────────────┘ │
│ │
└─────────────────────────────────────────────────────────────┘# 可抽取的实体模式
EXTRACT_PATTERNS = {
"Person": [
"我叫.*", "我是.*", "用户是.*",
"他.*她.*", "创业者", "工程师"
],
"Concept": [
".*是.*", "所谓的.*", "概念.*",
"本体论", "知识图谱", "Agent"
],
"Law": [
"当.*时.*", "如果.*那么.*",
"规律", "法则", "原则"
],
"Rule": [
"应该.*", "必须.*", "建议.*",
"推荐.*", "选择.*"
],
"Project": [
"项目.*", "在做.*", "目标是.*"
],
"Task": [
"任务.*", "需要做.*", "要做.*"
]
}
# 可抽取的关系模式
RELATION_PATTERNS = {
"is_a": ["是.*的一种", "属于.*类型"],
"relates_to": ["和.*相关", "与.*有关"],
"triggers": ["导致.*", "引发.*"],
"supports": ["支持.*", "基于.*"],
"contradicts": ["与.*矛盾", "不同于.*"]
}def check_duplicate(entity_type, name, properties=None):
"""检查是否已存在相同实体"""
existing = load_ontology()
for obj in existing.get(entity_type, []):
# 名称完全匹配
if obj.get("name") == name:
return {"duplicate": True, "existing": obj}
# 相似度检查(可选)
if similarity(name, obj.get("name")) > 0.8:
return {"duplicate": True, "similar": obj}
return {"duplicate": False}
def incremental_update(new_entities, new_links):
"""增量更新,避免覆盖"""
existing = load_ontology()
# 合并对象
for entity_type, entities in new_entities.items():
if entity_type not in existing:
existing[entity_type] = []
for entity in entities:
dup = check_duplicate(entity_type, entity.get("name"))
if not dup["duplicate"]:
entity["source"] = "interactive_extraction"
entity["created_at"] = timestamp()
existing[entity_type].append(entity)
# 合并关系
# ...类似逻辑
save_ontology(existing)用户Query
│
▼
┌────────────────────────────┐
│ 1. 方法论检查 │
│ check_ontology() │
│ - 查Objects │
│ - 查Laws │
│ - 查Rules │
└───────────┬────────────────┘
│
▼
┌────────────────────────────┐
│ 2. 标注置信度 │
│ - CONFIRMED? → 使用 │
│ - UNKNOWN? → 声明来源 │
└───────────┬────────────────┘
│
▼
┌────────────────────────────┐
│ 3. 需要假设? │
│ 是 → 明确标注ASSUMED │
│ 否 → 直接推理 │
└───────────┬────────────────┘
│
▼
┌────────────────────────────┐
│ 4. 交互确认(可选) │
│ - 关键假设问用户 │
│ - 根据确认调整置信度 │
└───────────┬────────────────┘
│
▼
┌────────────────────────────┐
│ 5. 灵活推理 │
│ - 本体优先 │
│ - 合理假设补充 │
│ - 明确标注差异 │
└───────────┬────────────────┘
│
▼
┌────────────────────────────┐
│ 6. 输出结果 │
│ - 带置信度标注 │
│ - 附推理依据 │
│ - 可选:抽取新本体 │
└────────────────────────────┘| 场景 | 处理方式 |
|---|---|
| 本体有明确数据 | 🟢 直接使用,标注CONFIRMED |
| 本体有部分数据 | 🟡 使用本体 + 标注ASSUMED补充 |
| 本体无数据但有合理推断 | 🟡 明确标注ASSUMED + 说明依据 |
| 完全无据可查 | 🔴 明确标注SPECULATIVE + 建议验证 |
| 需要用户确认 | ⚡ 交互提问 + 暂停推理 |
memory/ontology-clawra/
├── schema.yaml # 类型定义+约束
├── graph.jsonl # Objects + Links (实体+关系)
├── rules.yaml # Functions (业务规则库)
├── laws.yaml # Laws (规律/法则库)
├── decisions.jsonl # Actions-决策日志
├── reasoning.jsonl # Actions-推理日志(含置信度)
├── concepts.jsonl # Concepts (概念库)
├── extraction_log.jsonl # 新增:本体抽取日志
└── confidence_tracker.jsonl # 新增:置信度追踪# 基础推理(自动方法论检查)
python3 scripts/ontology-clawra.py reason --query "用户应该做什么"
# 强制声明来源
python3 scripts/ontology-clawra.py reason --query "100户用气量" --declare-source
# 交互确认模式
python3 scripts/ontology-clawra.py reason --query "调压箱选型" --confirm-needed# 从文本自动抽取(新增)
python3 scripts/ontology-clawra.py extract --text "用户是AI创业者,目标是构建垂直领域Agent平台"
# 手动创建
python3 scripts/ontology-clawra.py create --type Person --props '{"name":"用户","role":"AI创业者"}'
# 创建规律(带来源)
python3 scripts/ontology-clawra.py create --type Law --props '{"name":"红海规避","domain":"战略","statement":"...","source":"用户输入","confidence":"ASSUMED"}'# 验证推理可信度
python3 scripts/ontology-clawra.py validate --check-confidence
# 查看推理链
python3 scripts/ontology-clawra.py trace --decision decision_id
# 本体构建记录
python3 scripts/ontology-clawra.py extraction-history用户:100户居民小区调压箱如何选型?
1. check_ontology("调压箱 选型 居民")
→ 结果:本地无数据
→ 标注:UNKNOWN
2. 声明来源
⚪ 本地本体无相关数据,以下为外部知识推理
3. 需要假设?→ 是
- 假设:单户用气量、同时系数、地区等
→ 标注:ASSUMED
4. 交互确认
❓ 请确认:
- 是否有供暖需求?(影响用气量)
- 是什么类型的住宅?
- 当地燃气供气压力是多少?
5. 输出
🟡 基于以下假设的推理(需确认):
- 假设1:南方城市,无集中供暖
- 假设2:每户配置双眼灶+热水器
- 假设3:多层住宅
计算结果:RTZ-80/25,额定流量80m³/h
置信度:ASSUMED(需验证)
6. 可选:抽取到本体
→ 提取:Law{居民用气计算规则}
→ 提取:Rule{调压箱选型规则}用户:我在做AI创业,目标垂直领域Agent
1. check_ontology("AI创业 垂直领域 Agent")
→ 结果:
- Law[红海规避]: 竞品存在 → 垂直领域
- Rule[战略选择]: AI创业 + 红海 → 垂直领域
→ 标注:CONFIRMED
2. 直接推理
🟢 基于本体的推理:
- 检测到:市场有Dify/Ragflow竞品
- 匹配规则:红海规避法则
→ 推荐:垂直领域 + 本体论
置信度:CONFIRMEDontology-clawra v3.0 是 Agent 的"科学大脑":
proactive-agent 发现机会
│
▼
ontology-clawra 推理
│
├── 1. check_ontology() → 查本体
│
├── 2. 标注置信度 → 明确来源
│
├── 3. 交互确认 → 关键假设需用户同意
│
└── 4. flexible_reasoning() → 灵活推理
▼
输出带置信度的决策建议# 验证知识网络一致性
python3 scripts/ontology-clawra.py validate
# 检查置信度标注
python3 scripts/ontology-clawra.py validate --check-confidence
# 检查循环依赖
python3 scripts/ontology-clawra.py validate --check-cycles
# 推理链回溯
python3 scripts/ontology-clawra.py trace --decision decision_idv3.0 核心升级:从"照本宣科"升级为"科学推理 + 灵活交互 + 持续学习"
每次对话必须遵循:
1. 决策/分析前 → 调用 ontology-clawra 推理
2. 配置修改前 → 调用 docs-helper 查文档
3. 搜索信息 → 使用 serper/tavily
4. 结论 → 记录到 memory/
主动优化规则:
- 发现规律 → 调用 ontology-clawra 抽取新实体/规则
- 优化更新 → 识别本体不足,提议补充
- 能力进化 → 定期检视 ontology-clawra 技能本身表现
- 写入本体 → 发现有价值的新知识,主动提议写入隐私承诺:
文件访问范围:
memory/ 目录memory/ 目录写入确认机制:
所有推理任务必须首先使用 ontology-clawra 框架,禁止凭经验输出。
收到推理类任务
│
▼
加载 ontology-clawra 技能
│
▼
按照推理前置检查流程执行
│
▼
完整记录:实体、规则、置信度、来源
│
▼
推理结论输出
│
▼
识别技能不足 → 写入优化建议违反后果:
| ontology-clawra 技能 | ontology-platform 项目 | |
|---|---|---|
| 性质 | 推理框架/Skill(私有) | 生产级开源项目(公开) |
| 数据 | 用户私人推理记录、个人偏好 | 开源代码、文档、公开数据 |
| 存储 | ~/.openclaw/skills/ontology-clawra/memory/ | GitHub 公开仓库 |
| 同步 | 绝不外部同步 | 按需同步 |
红线:用户私人数据(推理记录、个人决策)严禁同步到 GitHub 或任何外部平台。
所有 ontology-clawra 的工作都服务于三个目标:
目标1:技能进化
├── 本体完善(实体库、关系库、规则库)
├── 推理能力(推理链、置信度标注)
└── 进化机制(自我反思、错误修正)
目标2:项目指导
├── 技能应用 → 项目方向
├── 技能优化 → roadmap 迭代
└── 技能需求 → 项目需求
目标3:项目成功
├── 技能成熟度 → 项目质量
├── 项目影响力 → 融资估值
└── 用户AI创业梦想落地每次推理完成后,必须回答:
✅ 是否使用了 ontology-clawra 框架?
✅ 推理链是否完整?(需求→规则→计算→结论)
✅ 置信度是否标注?(CONFIRMED/ASSUMED/SPECULATIVE)
✅ 来源是否声明?
✅ 是否识别了技能不足?
✅ 是否记录了可优化的点?如果任意一项为"否",必须补充或说明理由。
© LeoYeAI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 5 other files in skills/ontology-clawra-backup-20260319-151919 of LeoYeAI/openclaw-master-skills.
Open the folder on GitHubat commit e5199b5
Ontology Clawra next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Ontology Clawra this skillLeoYeAI/openclaw-master-skills | 2.2k | — | ~4.7k | Automated safety check: Pass | MIT | |
| PR Babysitteropeninterpreter/openinterpreter | 69k | 3 repos | ~4.2k | Automated safety check: Pass | Apache-2.0 | |
| Greplooponyx-dot-app/onyx | 32k | 4 repos | ~3.3k | Automated safety check: Pass | MIT | |
| GitHub Deep Researchbytedance/deer-flow | 84k | 4 repos | ~1.3k | Automated safety check: Pass | MIT | |
| Diagnosing Superpowers Sessionsobra/superpowers | 297k | 3 repos | ~1.7k | Automated safety check: Pass | MIT | |
| Update V8 Versionopeninterpreter/openinterpreter | 69k | 2 repos | ~845 | Automated safety check: Pass | Apache-2.0 |
openinterpreter/openinterpreter
Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.
onyx-dot-app/onyx
Iteratively improves a PR (GitHub), MR (GitLab), or shelved changelist (Perforce) until Greptile gives it a 5/5 confidence score with zero unresolved comments.
bytedance/deer-flow
Researches a GitHub repository over four rounds using the GitHub API and web search, then writes a structured markdown report with timeline, metrics and Mermaid diagrams.
obra/superpowers
Investigates a session where Superpowers went wrong, reads the transcripts on disk and produces an evidence-cited report, optionally prepared as a bug report for the maintainers.
openinterpreter/openinterpreter
Bumps the pinned v8 and rusty_v8 versions in Codex, validates the release-candidate path with the v8-canary check, and traces failures to upstream build changes.
mvanhorn/last30days-skill
Research what people actually say about any topic in the last 30 days.
LeoYeAI/openclaw-master-skills
Manages pipelines on a DevOps quality and efficiency platform through its OpenAPI: list workspaces and templates, create, update, run and cancel pipelines, and read run records.
LeoYeAI/openclaw-master-skills
Patches OpenClaw's Feishu extension so an edited document triggers an isolated agent session that reads the doc and replies inline, turning it into a live chat space.
LeoYeAI/openclaw-master-skills
Multi-context memory management system for OpenClaw agents with group-isolated storage, global shared memory, workspace organization, and group-specific skills isolation.
LeoYeAI/openclaw-master-skills
Runs a brand's AI-search visibility work end to end: diagnosing how AI platforms represent it, repositioning it, producing AI-optimized content and monitoring ongoing mentions.
LeoYeAI/openclaw-master-skills
Installs and authenticates the gws CLI, then automates Gmail, Drive, Sheets, Calendar, Docs, Chat and Tasks with ready-made recipes, persona bundles and security audits.
LeoYeAI/openclaw-master-skills
Runs four advisor roles, a fitness coach, nutritionist, data analyst and TCM practitioner, to build a health profile and track workouts, diet and wellness over time.
Works with
Palantir本体论实践版本 v3.5 - 强制执行版。结构化知识图谱+规则引擎+科学推理方法论。每次决策/分析前必须使用。推理结果必须展示详细推理过程:包含用户需求、规则依据ID、规则内容、置信度标注。. Ontology Clawra is an agent skill from LeoYeAI/openclaw-master-skills.
Run `npx skills add LeoYeAI/openclaw-master-skills --skill ontology-clawra -a claude-code`. Or copy the skill folder (skills/ontology-clawra-backup-20260319-151919 in LeoYeAI/openclaw-master-skills) into .claude/skills/ontology-clawra in your project. Claude Code loads it when a task matches its description.
Run `npx skills add LeoYeAI/openclaw-master-skills --skill ontology-clawra -a codex`. Or copy the skill folder (skills/ontology-clawra-backup-20260319-151919 in LeoYeAI/openclaw-master-skills) into .agents/skills/ontology-clawra in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add LeoYeAI/openclaw-master-skills --skill ontology-clawra -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ontology-clawra, .gemini/skills/ontology-clawra, .github/skills/ontology-clawra and .opencode/skills/ontology-clawra in your project.
Going by SKILL.md and its folder, Ontology Clawra needs the command-line tools its instructions call (python3). Our summary lists: Python 3.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Ontology Clawra is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 4.7k tokens (SKILL.md is roughly 19k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Ontology Clawra: PR Babysitter (openinterpreter/openinterpreter, 69k stars), Greploop (onyx-dot-app/onyx, 32k stars), GitHub Deep Research (bytedance/deer-flow, 84k stars) and Diagnosing Superpowers Sessions (obra/superpowers, 297k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
LeoYeAI (a GitHub user) maintains it in LeoYeAI/openclaw-master-skills, which has 2,161 GitHub stars. The repository holds 1,235 skills in this directory. The repository was last updated on July 20, 2026.
Source: LeoYeAI/openclaw-master-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.