Token导航 LogoToken导航TokenDH.com
研究检索权限需确认github未标认证来源可访问许可证需确认审计通过

status状态

Agent Skill

status 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 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。

总安装

11,407

周安装

457

GitHub Stars

318

下载量

3,693
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

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:status(状态)
来源仓库:https://github.com/boshu2/agentops
仓库路径:skills/status
安装命令:
npx skills add https://github.com/boshu2/agentops --skill status
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 npx skills 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。

skills.shnpx skills
npx skills add https://github.com/boshu2/agentops --skill status

简介

status 提供工作流总览面板,展示当前状态与下一步行动。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中掌握全局进展。
  • 并行整理 RPI、ratchet、任务列表等多源状态信息。
  • 支持 JSON 格式输出,便于与其他系统集成读取。
  • status 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

/status — Workflow Dashboard

Purpose: Single-screen overview of your current state. What am I working on? What happened recently? What should I do next?

YOU MUST EXECUTE THIS WORKFLOW. Do not just describe it.

CLI dependencies: bd, ao, gt — all optional. Shows what's available, skips what isn't.


Quick Start

/status              # Full dashboard
/status --json       # Machine-readable JSON output

Execution Steps

Step 1: Gather State (Parallel)

Run ALL of the following in parallel bash calls for speed:

Call 1 — RPI + Ratchet + Task State:

# Current ratchet phase
if [ -f .agents/ao/chain.jsonl ]; then
  tail -1 .agents/ao/chain.jsonl 2>/dev/null
else
  echo "NO_CHAIN"
fi

# Ratchet status via CLI
if command -v ao &>/dev/null; then
  ao ratchet status --json 2>/dev/null || echo "RATCHET_UNAVAILABLE"
  ao task-status --json 2>/dev/null || echo "TASK_STATUS_UNAVAILABLE"
fi

Call 2 — Beads / Epic State:

if command -v bd &>/dev/null; then
  echo "=== EPIC ==="
  bd list --type epic --status open 2>/dev/null | head -5
  echo "=== IN_PROGRESS ==="
  bd list --status in_progress 2>/dev/null | head -5
  echo "=== READY ==="
  bd ready 2>/dev/null | head -5
  echo "=== TOTAL ==="
  bd list 2>/dev/null | wc -l
else
  echo "BD_UNAVAILABLE"
fi

Call 3 — Knowledge Flywheel:

# Learnings count
echo "LEARNINGS=$(ls .agents/learnings/ 2>/dev/null | wc -l | tr -d ' ')"
echo "PATTERNS=$(ls .agents/patterns/ 2>/dev/null | wc -l | tr -d ' ')"
echo "PENDING=$(ls .agents/forge/ 2>/dev/null | wc -l | tr -d ' ')"

# Flywheel health + badge
if command -v ao &>/dev/null; then
  ao metrics flywheel status 2>/dev/null || echo "FLYWHEEL_UNAVAILABLE"
  ao badge 2>/dev/null || echo "BADGE_UNAVAILABLE"
fi

Call 4 — Recent Activity + Git:

# Recent sessions
if [ -d .agents/ao/sessions ]; then
  ls -t .agents/ao/sessions/*.md 2>/dev/null | head -3
else
  echo "NO_SESSIONS"
fi

# Recent council verdicts
ls -lt .agents/council/ 2>/dev/null | head -4

# Git state
echo "=== GIT ==="
git branch --show-current 2>/dev/null
git log --oneline -3 2>/dev/null
git status --short 2>/dev/null | head -5

Call 5 — Inbox:

if command -v gt &>/dev/null; then
  gt mail inbox 2>/dev/null | head -5
else
  echo "GT_UNAVAILABLE"
fi

Call 6 — Session Quality Signals:

if [ -f .agents/signals/session-quality.jsonl ]; then
  tail -10 .agents/signals/session-quality.jsonl
else
  echo "NO_SIGNALS"
fi

Step 2: Render Dashboard

Assemble gathered data into this format. Use Unicode indicators for visual clarity:

  • Pass/healthy: [PASS]
  • Warning/partial: [WARN]
  • Fail/missing: [FAIL]
  • Progress: [3/7] with bar ███░░░░
══════════════════════════════════════════════════
  Workflow Dashboard
══════════════════════════════════════════════════

RPI PROGRESS
  Phase: <current phase from chain.jsonl: research | plan | implement | validate | idle>
  Gate:  <last completed gate or "none">
  ─────────────────────────────────
  research ── plan ── implement ── validate
     <mark current position with arrow or highlight>

ACTIVE EPIC
  <epic title and ID, or "No active epic">
  Progress: <completed>/<total> issues  <progress bar>
  In Progress: <list in-progress issues, max 3>

READY TO WORK
  <top 3 unblocked issues from bd ready>
  <or "No ready issues — create work with /plan">

RECENT VALIDATIONS
  <last 3 council reports with verdict>
  <format: date  verdict  target>
  <or "No recent validations">

KNOWLEDGE FLYWHEEL
  Learnings: <count>  Patterns: <count>  Pending: <count>
  Health: <flywheel status or "ao not installed">
  Badge: <ao badge output or omit if unavailable>

TASK MATURITY
  <ao task-status summary: active tasks with CASS maturity levels, or omit if unavailable>

RECENT SESSIONS
  <last 3 session summaries with dates>
  <or "No session history">

GIT STATE
  Branch: <current branch>
  Recent: <last 3 commits, one-line>
  Changes: <uncommitted file count or "clean">

INBOX
  <message count or "No messages" or "gt not installed">

SESSION QUALITY SIGNALS
  <last 10 entries from .agents/signals/session-quality.jsonl as table>
  | Timestamp | Signal | Detail | Session |
  |-----------|--------|--------|---------|
  <parsed from JSON lines: .timestamp, .signal, .detail, .session>
  <or "No quality signals recorded." if file missing or empty>

──────────────────────────────────────────────────
SUGGESTED NEXT ACTION
  <state-aware suggestion — see Step 3>
──────────────────────────────────────────────────

QUICK COMMANDS
  /research     Deep codebase exploration
  /plan         Decompose epic into issues
  /pre-mortem   Validate plan before coding
  /implement    Execute a single issue
  /crank        Autonomous epic execution
  /validation   Full close-out and learnings
  /vibe         Targeted code review
══════════════════════════════════════════════════

Step 3: Suggest Next Action (State-Aware)

Evaluate state top-to-bottom. Use the FIRST matching condition:

PriorityConditionSuggestion
1No ratchet chain exists"Start with /quickstart or /research to begin a workflow"
2Research done, no plan"Run /plan to decompose research into actionable issues"
3Plan done, no pre-mortem"Run /pre-mortem to validate the plan before coding"
4Issues in-progress"Continue working: /implement <issue-id> or /crank for autonomous execution"
5Ready issues available"Pick up next issue: /implement <first-ready-id>"
6Uncommitted changes"Review recent work: /validation"
7Implementation done, no vibe"Run /validation for final close-out"
8Recent WARN/FAIL verdict"Address findings in <report-path>, then re-run /validation"
10Vibe passed, no post-mortem"Run /validation to complete closeout and extract learnings"
11Pending knowledge items"Promote learnings: ao pool list --status pending --json, then ao pool stage <id> and ao pool promote <id>"
12Clean state, nothing pending"All clear. Start with /research or /plan to find new work"

Step 4: JSON Output (--json flag)

If the user passed --json, output all dashboard data as structured JSON instead of the visual dashboard:

{
  "rpi": {
    "phase": "implement",
    "last_gate": "plan",
    "chain_entries": 3
  },
  "epic": {
    "id": "ag-042",
    "title": "Epic title",
    "progress": { "completed": 3, "total": 7, "in_progress": ["ag-042.2"] }
  },
  "ready_issues": ["ag-042.4", "ag-042.5"],
  "validations": [
    { "date": "2026-02-09", "verdict": "PASS", "target": "src/auth/" }
  ],
  "flywheel": {
    "learnings": 12,
    "patterns": 5,
    "pending": 2,
    "health": "healthy"
  },
  "sessions": [
    { "date": "2026-02-09", "file": "session-abc.md" }
  ],
  "git": {
    "branch": "main",
    "uncommitted_count": 3,
    "recent_commits": ["abc1234 fix: thing", "def5678 feat: other"]
  },
  "inbox": { "count": 0 },
  "session_quality_signals": [
    { "timestamp": "2026-03-31T14:22:00Z", "signal": "drift", "detail": "3 corrections in 5min", "session": "abc123" }
  ],
  "suggestion": {
    "priority": 5,
    "message": "Continue working: /implement ag-042.2"
  }
}

Render this with a single code block. No visual dashboard when --json is active.


Examples

Checking Status Mid-Epic

User says: /status

What happens:

  1. Agent runs 5 parallel bash calls to gather all state
  2. Agent reads ratchet chain showing "implement" phase
  3. Agent queries beads showing epic ag-042 with 3/7 issues completed
  4. Agent finds 2 in-progress issues and 4 ready issues
  5. Agent lists recent council verdict: PASS on src/auth/
  6. Agent checks flywheel showing 12 learnings, 5 patterns, 2 pending
  7. Agent renders dashboard with progress bars and suggests: "Continue working: /implement ag-042.2"

Result: Full single-screen dashboard showing mid-epic progress with actionable next step.

Status in Clean State

User says: /status

What happens:

  1. Agent gathers all state in parallel
  2. Agent finds no ratchet chain exists (.agents/ao/chain.jsonl missing)
  3. Agent finds no open epics or in-progress issues
  4. Agent shows clean git state, recent commits only
  5. Agent finds no recent validations
  6. Agent suggests: "All clear. Start with /research or /plan to find new work"

Result: Dashboard confirms clean slate, points user to workflow entry points.

Troubleshooting

ProblemCauseSolution
Shows "BD_UNAVAILABLE" or "AO_UNAVAILABLE"CLI tools not installed or not in PATHInstall missing tools: brew install bd or brew install ao. Skill gracefully degrades by showing available state only.
Ratchet phase shows stale dataOld chain.jsonl not cleaned upCheck timestamp of .agents/ao/chain.jsonl. If stale, delete it or run /validation to complete cycle and reset state.
Suggested action doesn't match intentState-aware rules didn't capture edge caseReview priority table in Step 3. May need to refine conditions. Use --json to inspect raw state and debug rule matching.
JSON output malformedParallel bash calls returned unexpected formatCheck each bash call individually. Ensure jq parsing works on actual data. Validate JSON structure with jq. before returning to user.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

38.27%
按下载量换算1,413

Claude

27.54%
按下载量换算1,017

Cursor

17.96%
按下载量换算663

Gemini CLI

8.61%
按下载量换算318

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

权限需确认

当前来源未能明确判断权限范围,默认进入异常复核队列。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills