pitch比稿技能
必赢逻辑引擎(Pitch Skill)— 专为广告/营销Agency的比稿竞标场景设计的AI影子智囊团。把资深策略总监脑子里的「玄学感悟」拆解为可计算的赢标逻辑。当用户需要在竞争性提案中赢下客户(多个供应商竞标、客户发RFP选Agency、评审团打分选方案)时使用此技能。6个Agent协作:Intake → In... Skill: pitch比稿技能 Owner: qomob Summary: 必赢逻辑引擎(Pitch Skill)— 专为广告/营销Agency的比稿竞标场景设计的AI影子智囊团。把资深策略总监脑子里的「玄学感悟」拆解为可计算的赢标逻辑。当用户需要在竞争性提案中赢下客户(多个供应商竞标、客户发RFP选Agency、评审团打分选方案)时使用此技能。6个Agent协作:Intake → In... Tags: latest:2.2.0 Version history: v2.2.0 | 2026-06-15T00:15:49.719Z | user - Major update: Introduces structured multi-agent execution, strict file loading protocols, and robust context management for better accuracy a
Rank
62
Safety
84
Downloads
1.0k
Updated
Oct 11, 2026
Version
2.2.0
Source
CLAWHUB
About
What it does, and when to use it.
Capability contract not published. No trust telemetry is available yet. 1K downloads reported by the source. Last updated 10/11/2026.
Avoid when
- Contract metadata is missing or unavailable for deterministic execution.
Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing
Public facts
Every fact links back to the source it came from.
- Vendor
- Clawhubvendor · observed Oct 11, 2026
- Protocol compatibility
- OpenClawcompatibility · observed Oct 11, 2026
- Adoption signal
- 1K downloadsadoption · observed Oct 11, 2026
- Latest release
- 2.2.0release · observed Jun 15, 2026
- Handshake status
- UNKNOWNsecurity
Install and run
Setup complexity: low.
clawhub skill install s1712hx0t1qgg41g18p2bb0d6583ms02:pitchskill- Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.
- Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.
Contract: missing
curl -s "https://www.xpersona.co/api/v1/agents/clawhub-qomob-pitchskill/snapshot"
Run-check
$0.02 USD1 measured facts are behind this paywall: success rate and latency, uptime and estimated cost, when not to use it, how to call it, benchmark scores.
Agents pay $0.02 in USDC. A card payment is $0.50, the smallest a card allows.
Documentation
CLAWHUB
145,900 characters of source documentation, loaded on request.
Extracted files
5 files captured from the source.
SKILL.md
---
name: pitch-skill
description: "必赢逻辑引擎(Pitch Skill)— 专为广告/营销Agency的比稿竞标场景设计的AI影子智囊团。把资深策略总监脑子里的「玄学感悟」拆解为可计算的赢标逻辑。当用户需要在竞争性提案中赢下客户(多个供应商竞标、客户发RFP选Agency、评审团打分选方案)时使用此技能。6个Agent协作:Intake → Information → Strategy → Decision → Expression → Delivery,覆盖Brief穿透与需求解构、决策者深度画像、竞标对手逻辑真空区推演、第一性原理策略推导、逻辑链自检、胜率计算、决策模拟、情绪引擎优化提案表达、AIGC具象化震撼Demo、Q&A压力训练。触发场景:比稿、竞标、pitch、提案竞标、agency pitch、RFP响应、招标方案、赢标策略、竞标方案、pitch deck准备、选代理商、换代理商、年度比稿、创意比稿、媒介比稿。也适用于客户要求正式presentation给管理层评审的场景。即使用户只说'帮我做个提案''有个比稿''要去pitch''客户要方案''准备比稿材料''要去竞标''帮我们赢下这个客户''怎么才能赢'等模糊表述,只要涉及向客户竞争性展示方案就应触发。不适用于:内部营销方案、融资路演、PPT美化、竞品调研、品牌定位、培训汇报等非竞争性场景。"
version: "2.2.0"
---
# Pitch Skill — 必赢逻辑引擎
你是比稿AI影子智囊团。甲方买的不是创意,买的是"解决问题的确定性"。目标只有一个:让用户赢下这场比稿。
## 三条铁律
贯穿所有Agent,违反任何一条会让系统沦为"内容生成工具":
1. **决策语言化** — 所有输出用 ROI / 风险 / 可执行性 / 决策影响 表达
2. **竞品推演** — 策略必须针对竞品弱点设计,找到"逻辑真空区"
3. **胜率评估** — 每个策略输出附带胜率评估 + 证据链
## 文件加载协议(必读)
**执行任何Agent前,必须严格按以下清单加载文件。不预加载未调用的Agent。**
```
加载清单(按Agent逐个加载):
Intake:
☐ agents/__init__.md — 注册表 + 降级策略 + 摘要协议
☐ agents/intake-agent.md
Information:
☐ agents/information-agent.md
Strategy:
☐ agents/strategy-agent.md
☐ references/strategy-frameworks.md
Decision:
☐ agents/decision-agent.md
☐ references/decision-engine.md
Expression:
☐ agents/expression-agent.md
☐ references/pitch-structure.md
Delivery:
☐ agents/delivery-agent.md
条件加载:
☐ references/bilingual-templates.md — 仅当用户使用英文提问时
```
**上下文管理规则:**
- Agent间传递结构化摘要(见 `agents/__init__.md` 的 Inter-Agent Handoff Protocol),不传递完整原始输出
- 每个Agent完成后,将输出压缩为摘要再传给下游,避免上下文膨胀
- 如果对话上下文接近模型上限,优先保留:策略路径 > 决策分析 > 情报细节
## Agent 索引
| Agent | 职责 | 定义文件 | 按需Reference |
|-------|------|----------|--------------|
| Intake 📋 | Brief结构化、作战卡 | [agents/intake-agent.md](agents/intake-agent.md) | — |
| Information 🔍 | 需求解构、决策者画像、竞品推演 | [agents/information-agent.md](agents/information-agent.md) | — |
| Strategy 🧠 | 第一性原理、逻辑链自检、策略路径 | [agents/strategy-agent.md](agents/strategy-agent.md) | [strategy-frameworks.md](references/strategy-frameworks.md) |
| Decision 🎯 | 决策模式、胜率计算、决策模拟 | [agents/decision-agent.md](agents/decision-agent.md) | [decision-engine.md](references/decision-engine.md) |
| Expression 🎤 | Pitch结构、情绪引擎、AIGC Demo、Q&A | [agents/expression-agent.md](agents/expression-agent.md) | [pitch-structure.md](references/pitch-structure.md) |
| Delivery 📦 | 交付打包、格式标准化 | [agents/delivery-agent.md](agents/delivery-agent.md) | — |
## 模式路由
| 模式 | 触发条件 | Agent调用链 |
|------|---------|------------|
| **Full** | 默认 | 全部6个Agent |
| **Preview** | 含"快速""preview""大致方案""先看看" | Intake → Information → Strategy(精简输出) |
| **Custom** | 用户指定Agent子集 | 自动补入最小依赖图,Intake不可跳过 |
| **Resume** | "从XX Agent继续" | 从指定Agent开始,从对话历史提取前置输出,缺失时提示用户补充 |
自定义编排依赖规则:Decision依赖Strategy,Expression依赖Decision。
## 降级与重试
- 每个Agent定义了降级策略(见各Agent文件和 `agents/__init__.md` Fallback Table)
- Agent输出不满足质量门控时,标注 ⚠️ 并继续,不阻断流水线
- 用户可在任意Checkpoint说"重做这个Agent"或"跳过这个Agent"
## Checkpoint
每个Agent完成后暂停等用户确认:
```
📌 Checkpoint [{序号}/6]: {Agent名} 已完成
{Markdown 摘要}
-README.md
# Pitch Skill — 必赢逻辑引擎
> 核心定位:AI影子智囊团,从"做方案"转向"造共识"。甲方买的不是创意,买的是"解决问题的确定性"。
>
> **Version 2.2.0** | SMM Level 3 (Validated)
## 一句话介绍
6个专业Agent协作,把资深策略总监脑子里的「玄学感悟」拆解为可计算的赢标逻辑。支持Brief穿透与需求解构、决策者深度画像、逻辑真空区推演、第一性原理策略推导、逻辑链自检、胜率计算、决策模拟、情绪引擎优化提案表达、AIGC具象化震撼Demo、Q&A压力训练。
## 三阶段作战逻辑
```
透视 → 重构 → 表达
| | |
挖掘 构建 制造
Brief 不可 高压迫感
背后的 替代的 场域+
Brief 策略 具象化震撼
```
## Agent 协作链
```
Intake Agent 📋 (项目启动/结构化)
→ Information Agent 🔍 (透视引擎 — 需求解构+决策者深度画像+竞品推演)
→ Strategy Agent 🧠 (重构引擎 — 第一性原理+逻辑链自检+策略路径)
→ Decision Agent 🎯 (决策引擎⭐核心壁垒)
→ Expression Agent 🎤 (表达引擎 — 情绪引擎+AIGC Demo+Q&A)
→ Delivery Agent 📦 (交付打包 — 6大标准交付物)
```
### 各Agent职责
| Agent | 职责 | 核心输出 |
|-------|------|---------|
| Intake 📋 | Brief结构化、隐性信号识别、质量门控 | 项目作战卡(Battle Card) |
| Information 🔍 | 需求解构(De-briefing)、决策者深度画像、竞品推演 | 真伪需求分类 + 逻辑真空区 |
| Strategy 🧠 | 第一性原理推导、逻辑链自检、策略路径 | 不可替代的策略路径 + Plan B/C |
| Decision 🎯 | 决策模式识别、胜率计算(含乘法下限)、决策模拟 | 胜率评估 + 证据链 + 优化建议 |
| Expression 🎤 | 情绪引擎、AIGC Demo、Q&A训练 | 8段式Pitch + 视觉Demo + 20题Q&A |
| Delivery 📦 | 交付打包、完整性检查、一致性校验 | 6大标准交付物 |
## 核心差异化
1. **需求解构(De-briefing)** — 穿透Brief表面,分离真痛点、伪需求、隐性需求
2. **第一性原理推导** — 从行业底层逻辑出发,否定平庸切入点
3. **逻辑链自检** — AI校验策略推导中是否有跳跃或想当然
4. **情绪引擎** — 逐段落评估情感冲击力,给出文案级优化建议
5. **AIGC具象化震撼** — 输出可直接使用的AI图像生成提示词,拉高竞争门槛
6. **胜率评估+证据链** — 区分"内容工具"和"赢标系统"的根本标志
## 执行模式
| 模式 | 触发条件 | Agent调用链 | Est. Cost |
|------|---------|------------|-----------|
| **Full** | 默认完整模式 | 全部6个Agent | ~$0.35-0.70 |
| **Preview** | 含"快速""preview""大致方案" | Intake → Information → Strategy(精简输出) | ~$0.15-0.30 |
| **Custom** | 用户指定Agent子集 | 自动补入最小依赖图 | 变化 |
| **Resume** | "从XX Agent继续" | 从指定Agent开始,复用前置输出 | 变化 |
成本估算基于 Claude Sonnet 级别模型。实际成本取决于Brief复杂度和交互轮数。
## 标准交付物
1. **Pitch Deck结构** — 内容逻辑版,每页含核心内容和演讲要点
2. **Strategy Doc** — 完整策略推导逻辑(含第一性原理+逻辑链自检报告)
3. **Q&A金句库** — 20个尖锐问题的30秒标准回答+节奏类型
4. **决策分析报告**⭐ — 决策模式 + 权力图谱 + 胜率 + 证据链 + 优化建议
5. **Win Rate评分** — 五维评分卡(含乘法下限保护)+ 优化路线图
6. **AIGC Demo提示词包**⭐ — 3-5个核心场景的AI图像生成提示词
## 文件结构
```
pitch-skill/
├── SKILL.md # 主文件(入口)
├── version.json # SSOT版本追踪 + 性能基线
├── baseline.json # 评估基线快照
├── .gitignore
├── agents/
│ ├── __init__.md # Lazy-loading注册表 + 降级策略 + 摘要协议 + 成本估算
│ ├── intake-agent.md # 项目启动引擎(含Brief质量门控)
│ ├── information-agent.md # 透视引擎(需求解构+决策者深度画像+竞品推演)
│ ├── strategy-agent.md # 重构引擎(第一性原理+逻辑链自检+策略路径)
│ ├── decision-agent.md # 决策引擎⭐(决策模式+胜率计算+模拟)
│ ├── expression-agent.md # 表达引擎(8段式Pitch+情绪引擎+AIGC Demo+Q&A)
│ └── delivery-agent.md # 交付引擎(完整性检查+一致性校验+6大交付物)
├── references/
│ ├── decision-engine.md # 决策引擎方法论(含乘法下限公式)
│ ├── pitch-structure.md # Pitch结构模板 + 情绪引擎模板 + AIGC Demo模板
│ ├── strategy-frameworks.md # 策略框架库 + 第一性原理方法论 + 逻辑链校验
│ └── bilingual-templates.md # 中英文术语对照与英文输出模板
├── evals/
│ ├── evals.json _meta.json
{
"ownerId": "kn72r3ww47r1qyfaf233sf7q1982rh5f",
"slug": "pitchskill",
"version": "2.2.0",
"publishedAt": 1781482549719
}references/bilingual-templates.md
# Bilingual Templates — 中英文术语对照与输出模板
When the user's input language is English, all Agent outputs switch to English. This file provides the English output templates and key term mappings.
## Key Term Glossary
| 中文 | English | Context |
|------|---------|---------|
| 比稿 | Competitive Pitch / Pitch | Agency selection process |
| 竞标 | Bid / Tender | Formal procurement |
| 提案 | Pitch / Proposal | Client presentation |
| 招标 | RFP (Request for Proposal) | Client procurement document |
| 作战卡 | Battle Card | Intake output |
| 隐性信号 | Hidden Signals | Brief subtext analysis |
| 情报层 | Information Engine | Intelligence gathering |
| 策略层 | Strategy Engine | Strategic planning |
| 决策层 | Decision Engine | Decision intelligence |
| 表达层 | Expression Engine | Pitch expression |
| 交付层 | Delivery Engine | Output packaging |
| 问题重构 | Problem Reframing | Three-layer reframing |
| 本质问题 | Essential Problem | Root cause identification |
| 洞察 | Insight | Consumer truth + brand intersection |
| 策略路径 | Strategy Path | Challenge → Insight → Idea → Framework → Impact |
| 风险对冲 | Risk Hedging | Conservative / Balanced / Aggressive versions |
| 决策模式 | Decision Mode | Safety / Political / Aggressive / Procurement |
| 权力图谱 | Power Graph | Stakeholder influence mapping |
| 胜率计算 | Win Probability | Five-dimension scoring |
| 决策模拟 | Decision Simulation | Mock pitch meeting |
| 竞品推演 | Shadow Pitch | Competitor strategy prediction |
| 逻辑空位 | Strategy Gap | Uncontested strategy space |
| 情绪曲线 | Emotion Curve | Pitch emotional rhythm |
| 黄金开场 | Opening Hook | First impression statement |
| Q&A压力训练 | Q&A Red Team | Hostile question preparation |
| 保守版/折中版/激进版 | Conservative / Balanced / Aggressive | Risk hedging versions |
| 决策者/影响者/否决者 | Decider / Influencer / Veto Holder | Stakeholder roles |
| 隐形决策者 | Hidden Decider | Off-table power player |
## English Output Templates
### Battle Card (Intake Agent)
```
Project Battle Card:
Client: {name} | Industry: {industry} | Stage: {Growth/Transition/Crisis/Maintenance}
Objective:
Type: {Growth/Brand/Transition/Crisis/Maintenance}
Primary Goal: {one sentence}
KPI Hints: {hinted KPI directions}
Constraints:
Budget: {amount or "Flexible"}
Timeline: Brief received {date} → Deadline {date} → Pitch {date} ({N} working days)
Channels: {specified channels}
Special Requirements: {any constraints}
Deliverables: {list}
Hidden Signals:
- {Signal}: {evidence from Brief} → {implication for strategy}
Competition:
Estimated Competitors: {number}
Likely Types: {competitor profiles}
Our Advantage: {core competitive edge}
Battle Card Summary:
One-line Strategy: {one sentence}
Must-Win Dimension: {dimension}
Avoid Dimension: {dimension}
Risk Level: {Low/Medium/High}
Recommended Approach: {Attack/Defend/Differentiate}
```
### Strategy Path (Strategy Agent)
```
Strategy Path:
CHALLENGE
{What is the market reality? Why is the status quo unsustainable?}
↓ INSIGHT
{What did we discover that others missed?}
↓ STRATEGIreferences/decision-engine.md
# 决策引擎方法论(Decision Engine Methodology) ## 目录 1. [决策模式识别](#决策模式识别) 2. [权力图谱构建](#权力图谱构建) 3. [胜率计算模型](#胜率计算模型) 4. [决策模拟方法](#决策模拟方法) 5. [优化策略库](#优化策略库) --- ## 决策模式识别 ### 四种决策模式 **Safety型决策** - 心理模型:决策者首要目标是"不犯错" - 信号词:稳健、可控、风险控制、保障、先例 - 评审特征:反复确认执行细节、要求案例 - 赢标策略:强调确定性、展示同类案例、提供风险兜底 - 避免事项:过于激进的主张、缺乏数据支撑的创意 **Political型决策** - 心理模型:决策是多方博弈的结果,每个stakeholder有自己的利益 - 信号词:多部门协作、联合评审、综合考量 - 评审特征:不同部门关注不同维度、存在意见分歧 - 赢标策略:方案设计成"多赢"格局、每个stakeholder都能找到自己的利益点 - 避免事项:只讨好一个人、忽略任何有否决权的角色 **Aggressive型决策** - 心理模型:决策者想要"大动作",害怕错过机会 - 信号词:突破、颠覆、重新定义、引领、先发 - 评审特征:对创新方案容忍度高、喜欢大叙事 - 赢标策略:大创意先行、数据辅助、展示远见 - 避免事项:过于保守、只讲执行不讲愿景 **Procurement型决策** - 心理模型:采购视角,追求性价比和标准化 - 信号词:性价比、服务承诺、SLA、资质、合规 - 评审特征:标准化评分表、权重明确、价格敏感 - 赢标策略:突出服务保障、资质背书、执行团队 - 避免事项:纯创意展示、不谈价格和服务 ### 决策模式判断矩阵 | 维度 | Safety | Political | Aggressive | Procurement | |------|--------|-----------|------------|-------------| | Brief语言风格 | 谨慎保守 | 平衡周全 | 进取前瞻 | 标准规范 | | 评审团组成 | 高管主导 | 多部门 | CEO/创始人参与 | 采购主导 | | 行业特征 | 金融/医药/国企 | 大型集团/外企 | 互联网/消费/新经济 | 政府/公用事业 | | 历史选择 | 偏大公司 | 偏综合型 | 偏创意型 | 偏低价 | --- ## 权力图谱构建 ### Stakeholder分类框架 每个评审团成员可以归入以下类别: 1. **核心决策者(Decider)**:最终拍板权,通常1人 2. **关键影响者(Influencer)**:能左右决策者,通常2-3人 3. **否决者(Veto Holder)**:可以一票否决,但不会主动选择 4. **隐形决策者(Hidden Decider)**:不在场但有决定性影响 5. **潜在盟友(Potential Ally)**:立场可能倾向我们 6. **需要说服(Need to Convince)**:立场不明确 ### 影响力关系建模 关系类型: - **影响(Influences)**:A的立场会影响B的判断 - **否决(Veto)**:A可以否决B的选择 - **同盟(Alliance)**:A和B通常持相同立场 - **对立(Opposition)**:A和B的立场通常相反 ### Stakeholder策略矩阵 | | 支持我们 | 中立 | 反对我们 | |---|---------|------|---------| | 高权力 | 强化盟友关系 | 重点转化对象 | 危险区域,需要绕道或改变策略 | | 中权力 | 利用其影响力 | 预防转向反对 | 减弱其影响力 | | 低权力 | 不投入太多精力 | 不投入太多精力 | 忽略 | --- ## 胜率计算模型 ### 五维评分框架 ``` WinRate = max( floor_multiplier × weighted_avg, weighted_avg ) weighted_avg = ( 策略匹配度 × 0.30 + 决策者匹配度 × 0.25 + 竞品差异化 × 0.20 + 执行可信度 × 0.15 + 关系/价格因素 × 0.10 ) floor_multiplier = min(五维评分) / 10 ``` **乘法下限保护:** 如果任一维度得分极低,胜率会被拉低到接近该维度水平。避免"策略满分但无法落地"的虚高胜率。 **策略匹配度(Strategy Fit)** - 我们的策略是否直接回应了客户的本质问题 - 评分依据:问题重构 → 策略路径的推导链是否严密 **决策者匹配度(Decision Maker Fit)** - 方案是否对核心决策者的胃口 - 评分依据:决策模式 + Persona + 权力图谱 **竞品差异化(Competitor Differentiation)** - 相对竞品的独特性 - 评分依据:Strategy Gap是否被有效占据 **执行可信度(Execution Credibility)** - 客户是否相信我们能落地 - 评分依据:案例、团队、方法论、时间线合理性 **关系/价格因素(Relationship & Price)** - 非方案因素 - 评分依据:现有关系深度、价格竞争力、行业口碑 ### 胜率区间解读 | 胜率区间 | 含义 | 行动 | |---------|------|------| | 80%+ | 强势领跑 | 保持,注意不犯错 | | 60-80% | 有竞争力但不确定 | 针对性优化Top 2风险 | | 40-60% | 势均力敌 | 需要重大调整,聚焦差异化 | | <40% | 劣势明显 | 需要根本性策略转向 | --- ## 决策模拟方法 ### 模拟角色生成 每个评审团角色生成: 1. **Persona**:基于Information Agent的决策者建模 2. **关注点**:基于角色职能和决策模式 3. **反应逻辑**:基于方案内容与关注点的匹配度 4. **相互影响**:基于权力图谱的关系 ### 模拟场景设计 1. **标准场景**:按Pitch结构顺序模拟各方反应 2. **压力场景**:模拟最尖锐的质疑 3. **意外场景**:模拟计划外的情况(如突然提出预算削减) ### 模拟输出解读 - **偏正面**:方案基本可行,优化细节即可 - **偏观望**:需要加强确定性和数据支撑 - **偏负面**:需要重大调整,可能是策略方向问题 --- ## 优化策略库 ### 按决策模式的优化策略 **Safety型优化:** - 增加"确定性表达":ROI承诺区间、分阶段交付、效果对赌 - 加强"落地案例":同行业成功案例、可量化的结果 - 添加"风险兜底":Plan B触发条件、止损方案 - 强调"团队稳定性":核心成员履历、
AionUi
Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!
activepieces
AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents
cherry-studio
AI productivity studio with smart chat, autonomous agents, and 300+ assistants.
CopilotKit
The Frontend for Agents & Generative UI. React + Angular
Machine-readable data
The same record, as JSON, for agents and crawlers.
{
"facts": [
{
"factKey": "vendor",
"category": "vendor",
"label": "Vendor",
"value": "Clawhub",
"href": "https://clawhub.ai/qomob/skills/pitchskill",
"sourceUrl": "https://clawhub.ai/qomob/skills/pitchskill",
"sourceType": "profile",
"confidence": "medium",
"observedAt": "2026-10-11T19:08:17.159Z",
"isPublic": true
},
{
"factKey": "protocols",
"category": "compatibility",
"label": "Protocol compatibility",
"value": "OpenClaw",
"href": "https://www.xpersona.co/api/v1/agents/clawhub-qomob-pitchskill/contract",
"sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-qomob-pitchskill/contract",
"sourceType": "contract",
"confidence": "medium",
"observedAt": "2026-10-11T19:08:17.159Z",
"isPublic": true
},
{
"factKey": "traction",
"category": "adoption",
"label": "Adoption signal",
"value": "1K downloads",
"href": "https://clawhub.ai/qomob/pitchskill",
"sourceUrl": "https://clawhub.ai/qomob/pitchskill",
"sourceType": "profile",
"confidence": "medium",
"observedAt": "2026-10-11T19:08:17.159Z",
"isPublic": true
},
{
"factKey": "latest_release",
"category": "release",
"label": "Latest release",
"value": "2.2.0",
"href": "https://clawhub.ai/qomob/pitchskill",
"sourceUrl": "https://clawhub.ai/qomob/pitchskill",
"sourceType": "release",
"confidence": "medium",
"observedAt": "2026-06-15T00:15:49.719Z",
"isPublic": true
},
{
"factKey": "handshake_status",
"category": "security",
"label": "Handshake status",
"value": "UNKNOWN",
"href": "https://www.xpersona.co/api/v1/agents/clawhub-qomob-pitchskill/trust",
"sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-qomob-pitchskill/trust",
"sourceType": "trust",
"confidence": "medium",
"observedAt": null,
"isPublic": true
}
],
"events": [
{
"eventType": "release",
"title": "Release 2.2.0",
"description": "- Major update: Introduces structured multi-agent execution, strict file loading protocols, and robust context management for better accuracy and scalability. - Added dedicated Agent definition files and references for all six functional modules: Intake, Information, Strategy, Decision, Expression, and Delivery. - Enforced per-Agent file loading with mandatory loading checklists, contextual handoff rules, and agent-specific fallback (降级) strategies. - Enhanced mode routing (Full, Preview, Custom, Resume) and clarified dependency chaining between agents. - Implemented checkpoint confirmation after each agent output, supporting user intervention, redo, or bypass. - Included a versioned regression testing protocol to ensure output quality and stability during upgrades.",
"href": "https://clawhub.ai/qomob/pitchskill",
"sourceUrl": "https://clawhub.ai/qomob/pitchskill",
"sourceType": "release",
"confidence": "medium",
"observedAt": "2026-06-15T00:15:49.719Z",
"isPublic": true
}
]
}Record generated Oct 11, 2026.
For crawlers
This page is free to read. The run-check above is the only paid part, and it answers HTTP 402 until it is paid. Everything else here is public.
- One record, as JSON: card, facts, snapshot, contract, trust.
- Every agent, one feed: /.well-known/ai-catalog.json
- What this site sells, and the price: /.well-known/x402
- Paid run-check: /api/v1/agents/clawhub-qomob-pitchskill/run-check
