← 返回 Skills 市场
william202404

Tob Poc War Room

作者 william202404 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
47
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install tob-poc-war-room
功能描述
POC war-room skill for ToB AI delivery. Use when a user needs to diagnose POC status, track pass rate, classify blockers, build a 48-hour closure plan, detec...
使用说明 (SKILL.md)

tob-poc-war-room

Use this skill to run a ToB POC like a war room, not a loose project tracker.

The goal is to convert POC facts into:

  • current risk level
  • blocker ownership
  • 48-hour closure plan
  • customer communication move
  • decision on whether to trigger poc-to-contract-closer

Do not output generic project-management templates. Every output must tie back to concrete POC signals.

Required Inputs

Ask for missing essentials only when they change the decision:

  • POC objective and acceptance criteria
  • current day or phase
  • pass rate or accepted test count
  • open issue list with severity, owner, age, and next action
  • customer feedback or silence duration
  • decision maker / champion status
  • procurement path status: unknown / tender / price comparison / single-source / framework agreement / renewal
  • competitor or parallel POC signal, if any

If inputs are incomplete, still produce a triage note and list the missing evidence.

Field Rules

Pass Rate

  • >=90%: closeable if no P0 blocker remains and acceptance evidence is concrete.
  • 70-89%: yellow; focus the next 48h on the smallest set of blockers that affect acceptance.
  • \x3C70%: red; do not talk contract yet unless customer scope changed.
  • Unknown pass rate: red-yellow; first action is to force a measurable acceptance baseline.

Use the 93% pass-rate pattern as a reference signal, not a promise.

Blockers

Classify each issue:

  • P0: blocks acceptance or executive demo.
  • P1: affects key user confidence but has workaround.
  • P2: cosmetic, edge case, or post-POC backlog.

48h rule:

  • P0 older than 48h without owner/action is red.
  • P1 older than 48h becomes yellow-red if it affects the champion's narrative.
  • P2 should not consume war-room attention unless customer names it.

Customer Silence

  • 0-2 days: normal, continue planned follow-up.
  • 3 days: proactive touchpoint required.
  • 7 days: danger signal; trigger executive/champion path and closing diagnosis.
  • Silence after successful demo is not neutral. Treat it as an unowned buying-process risk.

Procurement Path

ToB POC cannot move cleanly into closing if procurement path is unknown.

Check:

  • tender or formal bidding
  • price comparison
  • single-source justification
  • framework agreement call-off
  • renewal / expansion
  • direct purchase under threshold

If pass rate is high but procurement path is unknown, trigger poc-to-contract-closer with a buying-process gap instead of sending a quote.

Ownership

No output may contain an action without:

  • owner
  • next step
  • deadline
  • expected evidence

If owner is unknown, assign a role placeholder such as Tech owner, Sales owner, or Customer champion.

Trigger To Closing

Trigger poc-to-contract-closer when:

  • pass rate is >=90%, or
  • all P0 blockers are closed and customer has accepted the core value, or
  • the POC has entered final demo / recap / procurement discussion, or
  • customer silence appears after a successful POC and the buying process must be restarted.

Do not trigger closing when the POC is still proving basic feasibility.

Return From Closing

If poc-to-contract-closer finds unresolved P0, pass rate below closeable level, or missing acceptance evidence, return to tob-poc-war-room and run a 48-hour issue closure plan. Closing and war-room are a two-way loop, not a one-way handoff.

Terminology:

  • closeable: enough accepted POC evidence to start closing diagnosis.
  • contract-ready: closer confirms buyer path, procurement path, closing window, and issue treatment.
  • Do not treat closeable as contract-ready.

Output Format

## POC War-Room Triage

### 1. Status
- Phase:
- Risk: Green / Yellow / Red
- Pass rate:
- Main reason:

### 2. Blocker Table
| Issue | Level | Age | Owner | Next action | Deadline | Evidence |

### 3. Next 48 Hours
1.
2.
3.

### 4. Customer Move
- Message objective:
- Recommended message:
- Who sends it:
- When:

### 5. Escalation
- Escalate? yes/no
- Escalation target:
- Reason:

### 6. Closing Trigger
- Trigger `poc-to-contract-closer`? yes/no
- Reason:
- Inputs to pass forward:

### 7. Return Path
- If closer rejects readiness, next war-room action:

Acceptance Example

Input:

POC day 5. 14/15 tests accepted. One SSO issue open for 60h. Customer champion likes results, but procurement not involved. No customer reply for 3 days after demo.

Expected reasoning:

  • pass rate is green, but SSO age and customer silence create yellow risk.
  • next 48h must close SSO or document workaround.
  • Sales should restart champion/procurement path.
  • trigger poc-to-contract-closer with the risk flagged, because the POC is closeable but not contract-ready.

Boundaries

  • Do not invent customer names, project names, exact contract amount, or private data.
  • Do not promise conversion probability.
  • Do not call a POC successful if acceptance criteria are missing.
  • Do not bury a P0 blocker under a positive summary.
安全使用建议
Install this only in a ClawHub/Convex maintainer context where you trust the repo-local tools and understand the effects of moderation, PR publishing, deployment, and review commands. Use the confirmation steps, verify targets before writes, and consider running autoreview with --no-yolo or AUTOREVIEW_YOLO=0 if you do not want nested Codex review to bypass sandbox approvals.
能力评估
Purpose & Capability
The artifacts cover ClawHub moderation, PR maintenance, UI proof, autoreview, and Convex setup/migration/performance workflows; these capabilities match the stated purposes, though some workflows can affect accounts, PR comments, deployments, or local code review execution.
Instruction Scope
The instructions generally require explicit targets, reasons, confirmation before moderation writes, verification after changes, and use of official or repo-local tools; I found no hidden prompt override, deception, or unrelated behavior.
Install Mechanism
The skill content is normal Markdown/YAML plus one disclosed Bash helper under .agents/skills/autoreview/scripts/autoreview; I found no post-install hook, obfuscated payload, or automatic startup mechanism.
Credentials
Networked tools such as gh, bun, npx convex, and repo-local moderation/proof commands are expected for the workflows; the autoreview helper defaults to a full-access nested Codex review mode, which is disclosed and opt-out but should be used only in trusted repos.
Persistence & Privilege
No stealth persistence was found. Some intended workflows use auth tokens, long-running dev servers, PR publishing, or moderation privileges, but these are disclosed and tied to the skill purpose.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install tob-poc-war-room
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /tob-poc-war-room 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial public release.
元数据
Slug tob-poc-war-room
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Tob Poc War Room 是什么?

POC war-room skill for ToB AI delivery. Use when a user needs to diagnose POC status, track pass rate, classify blockers, build a 48-hour closure plan, detec... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 47 次。

如何安装 Tob Poc War Room?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install tob-poc-war-room」即可一键安装,无需额外配置。

Tob Poc War Room 是免费的吗?

是的,Tob Poc War Room 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Tob Poc War Room 支持哪些平台?

Tob Poc War Room 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Tob Poc War Room?

由 william202404(@william202404)开发并维护,当前版本 v1.0.0。

💬 留言讨论