Agent测试
- 作者仓库星标 3,406
- 作者更新于 实时读取
- 作者仓库 claude-octopus
- 领域
- 工程开发
- 兼容 Agent
-
- Claude Code
- Cursor
- Cline
- Codex
- Windsurf
- Gemini CLI
- +20
- 信任分
- 88 / 100 · 社区维护
- 作者 / 版本 / 许可
- @nyldn · 未声明 license
- Token 消耗评级
- 低消耗
- 接入复杂程度
- 需简单配置
- 是否需要外部 API Key
- 不需要
- 兼容的系统
- 未声明(默认跨平台)
- 底层运行要求
- 无特殊要求
- 文件与系统权限
-
- 只读
- 允许写入 / 修改
- Shell 执行
- 读取环境变量
- 网络行为
- 仅限本地
- 安装命令数
- 26 条
档案由构建时根据 SKILL.md 与安装命令自动衍生,可能与作者实际意图存在差异。
需要注意: 未限定 allowed-tools,默认拥有全部工具权限。
---
name: skill-status
description: Show where you are in the workflow and what to do next — use for progress checks and orientation…
category: 工程开发
runtime: 无特殊运行时
---
# skill-status 输出预览
## PART A: 任务判断
- 适用问题:代码实现、重构、调试或代码审查。
- 输入要求:目标材料、限制条件、期望输出和验收方式。
- 证据边界:围绕“Overview / When to Use / The Process”读取原文规则,不把推断写成作者承诺。
## PART B: 执行结果
- **01** 任务判断:确认你的需求是否属于代码实现、重构、调试或代码审查,并标出输入、限制和预期结果。
- **02** 执行计划:优先按“Overview / When to Use / The Process”拆成步骤,说明每一步会读取什么、修改什么、产出什么。
- **03** 交付结果:给出可复制的命令、文件改动、检查清单或内容草稿,并说明如何继续迭代。
- **04** 风险边界:结合 读取文件、写入/修改文件、执行终端命令、读取环境变量、主要在本地完成、通常不需要额外 API Key 给出执行前确认项。
## Running Rules
- 读取文件、写入/修改文件、执行终端命令、读取环境变量;主要在本地完成;通常不需要额外 API Key。
- 先小样例验证,再放大到真实任务。
- 交付时同时给结果、检查口径和下一步迭代建议。 原文出现了 `/octo` 这类斜杠命令;如果你的 Agent 支持命令触发,优先用命令开场,再补充目标和边界。
告诉 Agent 目标文件或材料、期望结果、不可改范围、是否允许联网或执行命令。本 Skill 的权限画像是:读取文件、写入/修改文件、执行终端命令、读取环境变量。
先用一个小任务确认它会围绕“Overview / When to Use / The Process”工作;涉及文件或命令时,先看 diff、日志、预览或测试结果。
检查最终产物是否包含明确结果、必要证据和下一步动作;如果输出泛泛而谈,就补充输入、边界和验收标准后重跑。
---
name: skill-status
description: Show where you are in the workflow and what to do next — use for progress checks and orientation…
category: 工程开发
source: nyldn/claude-octopus
---
# skill-status
## 什么时候使用
- 把工程方向的常用动作沉淀成 Agent 可调用的技能 适合处理工程开发场景下的代码实现、调试、重构、测试或代码审查,核心价值是把输入、判断、执行、验证和交付边界固定下来,避免 Agent 泛泛回答。 把任务拆成可执行、可检查、可继续迭代…
- 面向代码实现、重构、调试或代码审查,优先处理能明确输入、步骤和验收标准的工作。
## 需要提供什么
- 目标材料、目录范围、期望结果和不可改动内容。
- 是否允许联网、执行命令、读写文件或调用外部服务。
## 执行规则
- 围绕「Overview / When to Use / The Process」组织步骤,不把推断写成作者事实。
- 读取文件、写入/修改文件、执行终端命令、读取环境变量;主要在本地完成;通常不需要额外 API Key。
- 先跑小样例,确认结果可检查后再扩大任务范围。
## 输出要求
- 给出最终产物、关键证据、验证方式和下一步动作。
- 信息不足时标记 unknown,不编造命令、平台或依赖。 作者原文负责流程事实;仓库文件负责来源和命令;流狐只补充适用场景、限制和质量判断。
skill "skill-status" {
输入层 -> 用户目标 + 目标文件 + 禁止范围 + 验收标准
上下文层 -> Overview / When to Use / The Process
规则层 -> SKILL.md 触发条件 / 执行顺序 / 输出格式
运行层 -> 无特殊运行时 | 读取文件、写入/修改文件、执行终端命令、读取环境变量 | 主要在本地完成
安全层 -> 通常不需要额外 API Key + 小任务验证 + diff / 日志复核
输出层 -> 可复制结果 + 检查清单 + 下一步迭代
} Host: Codex CLI — This skill was designed for Claude Code and adapted for Codex. Cross-reference commands use installed skill names in Codex rather than
/octo:*slash commands. Use the active Codex shell and subagent tools. Do not claim a provider, model, or host subagent is available until the current session exposes it. For host tool equivalents, seeskills/blocks/codex-host-adapter.md.
Project Status Dashboard
Overview
Display current project status, roadmap progress, blockers, and intelligent next-action suggestions based on workflow state.
Core principle: Read state → Display dashboard → Route intelligently.
When to Use
Use this skill when user asks:
- "What's the status?" or "Show me progress"
- "Where am I in the workflow?"
- "What should I do next?"
- "What's the current phase?"
- "Are there any blockers?"
- "What have I been working on?" or "Summarize recent work"
- "Update project memory" or "Update CLAUDE.md"
Do NOT use for:
- Creating new projects (use /octo:embrace)
- Modifying state (use octo-state.sh write_state)
- Detailed phase planning (use flow-* skills)
The Process
Phase 1: Check Project Initialization
Step 1: Verify .octo/ Directory Exists
# Check if project is initialized
if [[ ! -d ".octo" ]]; then
echo "No project initialized"
exit 1
fi
If .octo/ does not exist:
## Project Status
**Status:** Not initialized
No Claude Octopus project found in this directory.
### Get Started
Run `/octo:embrace [your project description]` to initialize a new project and start the Double Diamond workflow.
**Example:**
/octo:embrace build a user authentication system
This will:
1. Create .octo/ directory structure
2. Initialize STATE.md, PROJECT.md, ROADMAP.md
3. Begin Discover phase (research and exploration)
Stop here - do not proceed to Phase 2.
Phase 2: Read Current State
Step 1: Execute octo-state.sh read_state
# Read current state from STATE.md
./scripts/octo-state.sh read_state
Expected output format:
schema=1.0
last_updated=2026-02-02T10:30:00Z
current_phase=2
current_position=define-requirements
status=in_progress
Step 2: Parse State Variables
Extract key-value pairs:
current_phase- Phase number (1-4)current_position- Specific position within phasestatus- Current workflow statuslast_updated- Last state modification timestamp
Phase 3: Read Roadmap
Step 1: Read ROADMAP.md
# Read roadmap for phase overview
cat .octo/ROADMAP.md
Step 2: Extract Phase Information
Parse ROADMAP.md to identify:
- Phase names (Discover, Define, Develop, Deliver)
- Phase descriptions
- Success criteria for each phase
- Dependencies between phases
Phase 4: Display Dashboard
Step 1: Build Status Dashboard
## Project Status
**Phase:** Phase {current_phase} - {phase_name}
**Position:** {current_position}
**Status:** {status}
**Last Updated:** {last_updated}
### Roadmap Progress
- [x] Phase 1: Discover - complete
- [ ] Phase 2: Define - in_progress <-- YOU ARE HERE
- [ ] Phase 3: Develop - not_started
- [ ] Phase 4: Deliver - not_started
### Current Phase Details
**Phase 2: Define (Grasp)**
- **Goal:** Clarify requirements and scope
- **Position:** {current_position}
- **Status:** {status}
### Blockers
{blockers or "None"}
### Suggested Next Action
{intelligent routing based on status}
Step 2: Determine Phase Completion Status
Map current_phase and status to completion markers:
| Phase | Status | Marker |
|---|---|---|
| 1 | complete | [x] Phase 1: Discover - complete |
| 2 | in_progress | [ ] Phase 2: Define - in_progress <-- YOU ARE HERE |
| 3 | not_started | [ ] Phase 3: Develop - not_started |
| 4 | not_started | [ ] Phase 4: Deliver - not_started |
Phase 5: Intelligent Routing
Step 1: Map Status to Suggestion
Use this routing table:
| Status | Suggestion |
|---|---|
ready |
"Run /octo:embrace [description] to start the workflow" |
planning |
"Continue planning in current phase. Use /octo:define to refine requirements." |
building |
"Continue implementation. Use /octo:develop to build features." |
in_progress |
"Continue with current phase. Check .octo/phases/phase{N}/ for details." |
blocked |
"Review blockers above. Use /octo:issues to track and resolve issues." |
complete |
"Phase complete. Proceed to next phase or run /octo:ship to finalize." |
complete_with_gaps |
"Phase complete with known gaps. Review .octo/ISSUES.md before proceeding." |
shipped |
"Project delivered! Review .octo/LESSONS.md for retrospective." |
paused |
"Project paused. Resume with /octo:embrace or check .octo/STATE.md for context." |
Step 2: Phase-Specific Routing
If status=in_progress, provide phase-specific guidance:
Phase 1 (Discover):
Continue research and exploration.
- Use `/octo:research [topic]` for multi-AI research
- Use `/octo:debate [question]` for decision support
Phase 2 (Define):
Continue requirements clarification.
- Use `/octo:prd` to write product requirements
- Use `/octo:define` to refine scope
Phase 3 (Develop):
Continue implementation.
- Use `/octo:develop` to build features
- Use `/octo:tdd` for test-driven development
- Use `/octo:review` for code quality checks
Phase 4 (Deliver):
Continue validation and delivery.
- Use `/octo:deliver` for final review
- Use `/octo:security` for security audit
- Use `/octo:ship` to finalize delivery
Example Outputs
Example 1: Project Not Initialized
## Project Status
**Status:** Not initialized
No Claude Octopus project found in this directory.
### Get Started
Run `/octo:embrace [your project description]` to initialize a new project and start the Double Diamond workflow.
**Example:**
/octo:embrace build a user authentication system
Example 2: Active Project in Define Phase
## Project Status
**Phase:** Phase 2 - Define (Grasp)
**Position:** define-requirements
**Status:** in_progress
**Last Updated:** 2026-02-02T10:30:00Z
### Roadmap Progress
- [x] Phase 1: Discover - complete
- [ ] Phase 2: Define - in_progress <-- YOU ARE HERE
- [ ] Phase 3: Develop - not_started
- [ ] Phase 4: Deliver - not_started
### Current Phase Details
**Phase 2: Define (Grasp)**
- **Goal:** Clarify requirements and scope
- **Position:** define-requirements
- **Status:** in_progress
### Blockers
None
### Suggested Next Action
Continue requirements clarification.
- Use `/octo:prd` to write product requirements
- Use `/octo:define` to refine scope
- Check `.octo/phases/phase2/` for detailed plans
Example 3: Blocked Project
## Project Status
**Phase:** Phase 3 - Develop (Tangle)
**Position:** implement-auth
**Status:** blocked
**Last Updated:** 2026-02-02T14:15:00Z
### Roadmap Progress
- [x] Phase 1: Discover - complete
- [x] Phase 2: Define - complete
- [ ] Phase 3: Develop - blocked <-- YOU ARE HERE
- [ ] Phase 4: Deliver - not_started
### Current Phase Details
**Phase 3: Develop (Tangle)**
- **Goal:** Implement features
- **Position:** implement-auth
- **Status:** blocked
### Blockers
- Missing OAuth provider credentials
- Database schema not finalized
- API rate limiting not configured
### Suggested Next Action
Review blockers above. Use `/octo:issues` to track and resolve issues.
**To unblock:**
1. Configure OAuth credentials in .env
2. Finalize database schema with `/octo:define`
3. Set up rate limiting configuration
Phase 6: Recent Activity Summary (Cross-Session)
When the user asks "what have I been working on", "summarize recent work", or "update project memory", generate a cross-session activity summary.
Step 1: Gather Recent Activity
# Recent git commits (last 7 days or last 20 commits)
git log --oneline --since="7 days ago" --no-merges 2>/dev/null | head -20
# Recent tags/releases
git tag --sort=-creatordate | head -5
# Recent branches worked on
git branch --sort=-committerdate | head -5
# Recent orchestration results (if any)
ls -lt ~/.claude-octopus/results/ 2>/dev/null | head -10
Step 2: Summarize Activity
Build a concise summary grouped by theme:
## Recent Activity (Last 7 Days)
### Commits
- [theme 1]: brief summary of related commits
- [theme 2]: brief summary of related commits
### Releases
- v8.10.0 - Gemini CLI headless fix
- v8.9.0 - Contextual Codex model routing
### Active Branches
- main (current)
### Orchestration Sessions
- [count] workflows executed, [count] synthesis files generated
Step 3: Suggest CLAUDE.md Updates
If the recent activity reveals patterns not captured in CLAUDE.md, suggest specific additions:
### Suggested CLAUDE.md Updates
Based on recent activity, consider adding:
- [specific suggestion based on new patterns, conventions, or decisions]
- [specific suggestion based on new tooling or workflow changes]
Only suggest updates that reflect durable project knowledge (conventions, architecture decisions, provider configs) — NOT transient status like "currently working on X".
Integration with Other Skills
With flow-* skills
User asks "what's next?"
→ skill-status shows current phase
→ User runs /octo:develop to continue
With skill-task-management
User asks "show status"
→ skill-status displays dashboard
→ skill-task-management shows active todos
With /octo:embrace
User asks "status" but no .octo/ exists
→ skill-status suggests /octo:embrace
→ User initializes new project
Best Practices
1. Always Check for .octo/ First
Good:
if [[ ! -d ".octo" ]]; then
echo "Not initialized"
exit 1
fi
Poor:
# Assume .octo/ exists and fail later
cat .octo/STATE.md
2. Use octo-state.sh for State Reading
Good:
./scripts/octo-state.sh read_state
Poor:
# Parse STATE.md manually
grep "Current Phase" .octo/STATE.md
3. Provide Actionable Next Steps
Good:
Continue with `/octo:develop` to implement features.
Check `.octo/phases/phase3/` for implementation plan.
Poor:
You're in phase 3.
Red Flags - Don't Do This
| Action | Why It's Wrong |
|---|---|
| Modify STATE.md directly | Use octo-state.sh write_state instead |
| Skip .octo/ existence check | Will fail with confusing errors |
| Show status without next action | User doesn't know what to do |
| Hardcode phase names | Read from ROADMAP.md for accuracy |
| Ignore blockers | User needs to know what's blocking progress |
Quick Reference
| User Input | Action Required |
|---|---|
| "status" | Check .octo/ → Read state → Display dashboard |
| "what's next" | Read state → Route based on status/phase |
| "where am I" | Display current phase and position |
| "show progress" | Display roadmap with completion markers |
| "any blockers" | Extract and display blockers from STATE.md |
| "what have I been working on" | Git log + results → Cross-session activity summary |
| "update project memory" | Activity summary → Suggest CLAUDE.md additions |
The Bottom Line
Status check → Read state + roadmap → Display dashboard + intelligent routing
Otherwise → User doesn't know where they are or what to do next
Check initialization. Read state. Display clearly. Route intelligently.
先判断是否适合
作者设计意图
作者的方法与取舍
边界和复核