Token导航 LogoToken导航TokenDH.com
研究检索执行命令github未标认证来源可访问许可证需确认审计提醒

objectiveobjective 搜索

Agent Skill

objective 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

1,560

周安装

65

GitHub Stars

81

下载量

520
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:objective(objective 搜索)
来源仓库:https://github.com/dagster-io/erk
仓库路径:skills/objective
安装命令:
npx skills add https://github.com/dagster-io/erk --skill objective
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/dagster-io/erk --skill objective

简介

objective 搜索用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中快速定位候选结果。

  • 适用于需要根据关键词或任务场景进行信息检索的场景,如目标设定、任务规划等。
  • 通过 GitHub 仓库安装,使用 npx skills add 命令添加技能,需确认权限和维护状态。
  • 使用前应检查是否会触发联网、命令执行或文件读写,确保符合项目安全策略。
  • objective 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Objective Skill

Overview

Objectives are coordination documents for goals requiring multiple plans/PRs to complete. Unlike erk-prs (single executable implementations), objectives track progress across related work and capture lessons learned along the way.

Scope range:

  • Small: Feature requiring 2-3 related PRs
  • Medium: Refactor spanning several plans
  • Large: Long-running strategic direction emitting many plans

Objective vs Erk-Plan

AspectErk-PlanObjective
PurposeSingle executable implementationCoordinate 2+ related plans/PRs
ScopeOne PR or tightly-coupled changeMultiple plans toward coherent goal
BodyMachine-parseable metadataHuman-readable markdown
CommentsSession context dumpsAction logs + lessons
Labelerk-prerk-objective
Toolingerk pr dispatch/implementManual updates via comments

Key Design Principles

  1. Human-first - Plain markdown, no machine-generated metadata
  2. Incremental capture - Each action gets its own comment
  3. Lessons as first-class - Every action comment includes lessons learned
  4. Clear roadmap - Status visible at a glance in the body
  5. Body stays current via reconciliation - After every PR landing, agents audit prose sections against what was actually implemented and correct stale information
  6. Steelthread-first - Each phase starts with minimal vertical slice proving the concept works
  7. One PR per phase - Each phase is sized for a coherent single PR
  8. Always shippable - System remains functional after each merged PR
  9. Body is source of truth - Body always contains complete current state; comments are the changelog
  10. Two-step for all changes - Every addition (context, decisions, phases) gets a comment AND body update
  11. Context over code - Provide references to patterns, not prescriptive implementations
  12. Session handoff ready - Body should be self-contained for any session to pick up and implement

Quick Reference

Creating an Objective

gh issue create --title "Objective: [Title]" --label "erk-objective" --body "$(cat <<'EOF'
# Objective: [Title]

> [1-2 sentence summary]

## Goal

[What success looks like - concrete end state]

## Design Decisions

1. **[Decision name]**: [What was decided]

## Roadmap

### Phase 1: [Name] Steelthread (1 PR)

Minimal vertical slice proving the concept works.

| Node | Description | Status | PR |
|------|-------------|--------|-----|
| 1.1 | [Minimal infrastructure] | pending | |
| 1.2 | [Wire into one command] | pending | |

**Test:** [End-to-end acceptance test for steelthread]

### Phase 2: Complete [Name] (1 PR)

Fill out remaining functionality.

| Node | Description | Status | PR |
|------|-------------|--------|-----|
| 2.1 | [Extend to remaining commands] | pending | |
| 2.2 | [Full test coverage] | pending | |

**Test:** [Full acceptance criteria]

EOF
)"

Logging an Action

Post an action comment after completing work. See format.md for full template.

gh issue comment <issue-number> --body "$(cat <<'EOF'
## Action: [Brief title]
**Date:** YYYY-MM-DD | **PR:** #123 | **Phase/Node:** 1.2
### What Was Done
### Lessons Learned
### Roadmap Updates
EOF
)"

After posting, update the issue body (roadmap statuses, reconcile stale prose sections).

Updating Node Details

Update node description, slug, or reason using update-objective-node:

# Update description
erk exec update-objective-node 8470 --node 1.3 --description "Revised description"

# Set slug
erk exec update-objective-node 8470 --node 1.3 --slug "revised-slug"

# Skip with comment
erk exec update-objective-node 8470 --node 1.3 --status skipped --comment "Superseded by new approach"

# Combine multiple updates
erk exec update-objective-node 8470 --node 2.1 --description "New desc" --slug "new-slug" --status planning

Adding Nodes

Add new nodes to an existing objective's roadmap:

# Add to existing phase (auto-assigns next ID, e.g., 1.4)
erk exec add-objective-node 8470 --phase 1 --description "Clean up dead code"

# With explicit slug and dependencies
erk exec add-objective-node 8470 --phase 2 \
  --description "Integration tests" \
  --slug integration-tests \
  --depends-on 2.1 --depends-on 2.2

# With comment for adding
erk exec add-objective-node 8470 --phase 1 \
  --description "Handle edge case" \
  --comment "Discovered during re-evaluation"

Updating Node Status

Always use the programmatic command — never manually edit YAML or prose tables:

erk exec update-objective-node <issue> --node <id> --status <status>
erk exec update-objective-node <issue> --node <id> --pr '#1234'

Then validate: erk objective check <issue-number>

Viewing an Objective

View an objective's dependency graph, dependencies, and next node:

# CLI
erk objective view <issue-number>

# Slash command (in-session)
/local:objective-view <issue-number>

Spawning an Erk-Plan

To implement a specific roadmap node, create an erk-pr that references the objective:

erk pr create --file plan.md --title "Implement [node description]"

Workflow Summary

  1. Create objective - When starting multi-plan work
  2. Inspect progress - View dependency graph and next node
  3. Log actions - After completing each significant piece of work
  4. Update body - Keep roadmap status current, reconcile stale prose after each PR landing
  5. Spawn erk-prs - For individual implementation nodes
  6. Close - When goal achieved or abandoned (proactively ask when all nodes done)

Resources

references/

  • format.md - Complete templates, examples, and update patterns
  • workflow.md - Creating objectives, spawning plans, steelthread structuring
  • updating.md - Quick reference for the two-step update workflow
  • closing.md - Closing triggers and procedures

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

33.83%
按下载量换算176

Claude

30.6%
按下载量换算159

Cursor

18.03%
按下载量换算94

Gemini CLI

9.11%
按下载量换算47

安全审计

Gen Agent Trust Hub

通过

Socket

可疑

Snyk

可疑

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add https://github.com/dagster-io/erk --skill objective 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 warning/failed 结果,不能写成本站确认安全。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