Token导航 LogoToken导航

Agent Skill · 研究检索

mini-piv(迷你 piv)

mini-piv 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

mini-piv 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为执行命令;许可证为 MIT-0,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 2,310。

总安装 63,523周安装 2,545GitHub Stars 公开资料未说明下载 20,564核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.1.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:mini-piv(迷你 piv)
来源仓库:https://github.com/smokealot420/mini-piv
安装命令:
openclaw skills install mini-piv
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install mini-piv

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

轻量级 PIV 工作流程 - 发现驱动的特征构建器。无需 PRD。提出快速问题,生成 PRP,通过验证循环执行。适用于想要跳过 PRD 仪式的中小型功能。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install mini-piv;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要执行命令,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

mini-piv 怎么安装?

在终端执行 openclaw skills install mini-piv。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

mini-piv 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

mini-piv 需要什么权限?

权限级别是执行命令。安装流程涉及命令执行,可能通过 openclaw skills install mini-piv 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

mini-piv 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

mini-piv 用的人多吗?

公开数据是累计安装 63,523 次,近一周安装 2,545 次,下载 20,564 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

mini-piv 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。

先想清楚的地方

  • 需要执行命令,装到常用环境前要先评估影响面。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 不能接受 Skill 在你机器上执行命令的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
mini-piv

迷你 piv · 本页

63,523—执行命令MIT-0有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
mini-piv
description
Lightweight PIV workflow - discovery-driven feature builder. No PRD needed. Asks quick questions, generates PRP, executes with validation loop. For small-to-medium features when you want to skip PRD ceremony.
user-invocable
true
disable-model-invocation
true
metadata
{"openclaw":{"emoji":"zap","homepage":"https://github.com/SmokeAlot420/ftw","requires":{"bins":["git"]},"os":["darwin","linux"]}}

Mini PIV Ralph - Lightweight Feature Builder

Arguments: $ARGUMENTS

Parse arguments:

FEATURE_NAME = $ARGUMENTS[0] or null (will ask user during discovery)
PROJECT_PATH = $ARGUMENTS[1] or current working directory

Philosophy: Quick & Quality

"When you just want to build something without writing a PRD first."

Same quality pipeline (Execute → Validate → Debug), but starts from a quick conversation instead of a PRD.

You are the orchestrator - stay lean, spawn fresh sub-agents for heavy lifting.

Sub-agent spawning: Use the sessions_spawn tool to create fresh sub-agent sessions. Each spawn is non-blocking — you'll receive results via an announce step. Wait for each agent's results before proceeding to the next step.


Required Reading by Role

RoleInstructions
OrchestratorThis file only
Research Agent{baseDir}/references/codebase-analysis.md + {baseDir}/references/generate-prp.md
Executor{baseDir}/references/piv-executor.md + {baseDir}/references/execute-prp.md
Validator{baseDir}/references/piv-validator.md
Debugger{baseDir}/references/piv-debugger.md

Visual Workflow

┌──────────────────────────────────────────────────────────┐
│ 1. DISCOVERY → Ask 3-5 questions                          │
│ 2. RESEARCH & PRP → Codebase analysis + PRP generation    │
│ 3. EXECUTE → Implement PRP                                │
│ 4. VALIDATE → PASS / GAPS_FOUND / HUMAN_NEEDED            │
│ 5. DEBUG LOOP → Fix gaps (max 3x)                         │
│ 6. COMMIT → feat(mini): {description}                     │
└──────────────────────────────────────────────────────────┘

Step 1: Discovery Phase

1a. Determine Feature Name

If not provided: ask user or infer from context. Normalize to kebab-case.

1b. Check for Existing PRP

ls -la PROJECT_PATH/PRPs/ 2>/dev/null | grep -i "mini-{FEATURE_NAME}"

If exists, ask: "Overwrite, rename, or skip to execution?"

1c. Ask Discovery Questions

Present in a single conversational message:

I've got a few quick questions so I can build this right:

1. **What does this feature do?** Quick rundown.
2. **Where in the codebase does it live?** Files, folders, components?
3. **Any specific libraries, patterns, or existing code to follow?**
4. **What does "done" look like?** 1-3 concrete success criteria.
5. **Anything explicitly OUT of scope?**

Adapt for feature type (UI, API, contracts, integrations).

1d. Structure Discovery Answers

feature:
  name: {FEATURE_NAME}
  scope: {Q1}
  touchpoints: {Q2}
  dependencies: {Q3}
  success_criteria: {Q4}
  out_of_scope: {Q5}

Step 2: Research & PRP Generation

Spawn a fresh sub-agent using sessions_spawn:

MINI PIV: RESEARCH & PRP GENERATION
====================================

Project root: {PROJECT_PATH}
Feature name: {FEATURE_NAME}

## Discovery Input
{paste structured YAML}

## Step 1: Codebase Analysis
Read {baseDir}/references/codebase-analysis.md for the process.
Save to: {PROJECT_PATH}/PRPs/planning/mini-{FEATURE_NAME}-analysis.md

## Step 2: Generate PRP (analysis context still loaded)
Read {baseDir}/references/generate-prp.md for the process.

### Discovery → PRP Translation
| Discovery | PRP Section |
|-----------|-------------|
| Scope (Q1) | Goal + What |
| Touchpoints (Q2) | Implementation task locations |
| Dependencies (Q3) | Context YAML, Known Gotchas |
| Success Criteria (Q4) | Success Criteria + Validation |
| Out of Scope (Q5) | Exclusions in What section |

Use template: PRPs/templates/prp_base.md
Output to: {PROJECT_PATH}/PRPs/mini-{FEATURE_NAME}.md

Do BOTH steps yourself. DO NOT spawn sub-agents.

Wait for completion.


Step 3: Spawn EXECUTOR

Spawn a fresh sub-agent using sessions_spawn:

EXECUTOR MISSION - Mini PIV
============================

Read {baseDir}/references/piv-executor.md for your role.
Read {baseDir}/references/execute-prp.md for the execution process.

PRP: {PROJECT_PATH}/PRPs/mini-{FEATURE_NAME}.md
Project: {PROJECT_PATH}

Follow: Load PRP → Plan Thoroughly → Execute → Validate → Verify
Output EXECUTION SUMMARY.

Validation Sizing Decision

Before spawning a full validator, assess:

  • <5 files changed, <100 lines, no external APIs → Quick validation (review changes yourself as orchestrator)
  • Otherwise → Spawn full validator sub-agent (Step 4)

Step 4: Spawn VALIDATOR

Spawn a fresh sub-agent using sessions_spawn:

VALIDATOR MISSION - Mini PIV
=============================

Read {baseDir}/references/piv-validator.md for your process.

PRP: {PROJECT_PATH}/PRPs/mini-{FEATURE_NAME}.md
Project: {PROJECT_PATH}
Executor Summary: {SUMMARY}

Verify ALL requirements independently.
Output VERIFICATION REPORT with Grade, Checks, Gaps.

Process result: PASS → commit | GAPS_FOUND → debugger | HUMAN_NEEDED → ask user


Step 5: Debug Loop (Max 3 iterations)

Spawn a fresh sub-agent using sessions_spawn:

DEBUGGER MISSION - Mini PIV - Iteration {I}
============================================

Read {baseDir}/references/piv-debugger.md for your methodology.

Project: {PROJECT_PATH}
PRP: {PROJECT_PATH}/PRPs/mini-{FEATURE_NAME}.md
Gaps: {GAPS}
Errors: {ERRORS}

Fix root causes. Run tests after each fix.
Output FIX REPORT.

After debugger: re-validate → PASS (commit) or loop (max 3) or escalate.


Step 6: Smart Commit

cd PROJECT_PATH && git status && git diff --stat
git add -A
git commit -m "feat(mini): implement {FEATURE_NAME}

- {bullet 1}
- {bullet 2}

Built via Mini PIV Ralph

Built with FTW (First Try Works) - https://github.com/SmokeAlot420/ftw"

Completion

## MINI PIV RALPH COMPLETE

Feature: {FEATURE_NAME}
Project: {PROJECT_PATH}

### Artifacts
- PRP: PRPs/mini-{FEATURE_NAME}.md
- Analysis: PRPs/planning/mini-{FEATURE_NAME}-analysis.md

### Implementation
- Validation cycles: {N}
- Debug iterations: {M}

### Files Changed
{list}

All requirements verified and passing.

Error Handling

  • Executor BLOCKED: Ask user for guidance
  • Validator HUMAN_NEEDED: Ask user for guidance
  • 3 debug cycles exhausted: Escalate with persistent issues list

Sub-Agent Timeout/Failure

When a sub-agent times out or fails:

  1. Check for partial work (files created, tests written)
  2. Retry once with a simplified, shorter prompt
  3. If retry fails, escalate to user with what was accomplished

Quick Reference

ScenarioUse This
Small/medium feature, no PRDMini PIV
Large feature with phasesFull PIV (/piv)

File Naming

PRPs/mini-{feature-name}.md                  # PRP
PRPs/planning/mini-{feature-name}-analysis.md # Analysis

平台分布

OpenClaw

75.15%
按下载量换算15,454

安全审计

VirusTotal

可疑

ClawScan

通过

Static analysis

未展示

权限和风险

执行命令

安装流程涉及命令执行,可能通过 openclaw skills install mini-piv 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills