npx skills add https://github.com/incept5/eve-skillpacks --skill eve-plan-implementation将计划文档转化为 Eve 任务,并行化工作,并通过任务阶段和依赖关系驱动评审/验证。
编排模型:根史诗是 编排器 —— 它负责规划、委派和协调,但不亲自执行繁重的工作。阶段任务是 子编排器,它们将一个阶段分解为多个任务。任务作业是 工作者 —— 每个任务接收一个自包含的描述并独立执行,无法访问父级上下文。
阅读计划文档并提取阶段、交付成果和阻塞项。
如果存在,阅读 AGENTS.md 以获取仓库特定规则。
获取当前任务上下文:
eve job current --json
保持轻量级:编排器仅读取足够的信息来规划分解。将深入分析(阅读大型文件、探索代码)委托给工作者任务。
如果根任务不存在,则创建一个:
eve job create \
--project $EVE_PROJECT_ID \
--description "实施 <计划名称>" \
--review human \
--phase backlog
如果根任务已存在,则将其用作编排器。根史诗从不执行实施工作 —— 它创建阶段任务、建立依赖关系并等待。
为每个计划阶段创建一个子任务。每个阶段任务充当子编排器:将其范围分解为任务作业并进行协调。
Translate a plan document into Eve jobs, parallelize work, and drive review/verification through job phases and dependencies.
Orchestration model : The root epic is the orchestrator — it plans, delegates, and coordinates but does not execute heavy work itself. Phase jobs are sub-orchestrators that break a phase into tasks. Task jobs are workers — each one receives a self-contained description and executes independently with no access to the parent's context.
Read the plan doc and extract phases, deliverables, and blockers.
If present, read AGENTS.md for repo-specific rules.
Fetch current job context:
eve job current --json
Stay lightweight: the orchestrator reads just enough to plan the breakdown. Delegate deep analysis (reading large files, exploring code) to worker jobs.
If the root job does not exist, create one:
eve job create \
--project $EVE_PROJECT_ID \
--description "Implement <plan name>" \
--review human \
--phase backlog
广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
eve job create \
--project $EVE_PROJECT_ID \
--parent $EVE_JOB_ID \
--description "阶段:<名称>。交付成果:<工件/结果>" \
--phase ready
添加依赖关系,使父任务等待每个阶段:
eve job dep add $EVE_JOB_ID $PHASE_JOB_ID --type waits_for
将每个阶段拆分为 2-6 个具有明确交付成果的原子任务。
如果一个阶段只有一个任务,则直接在该阶段任务中执行,而不是创建子任务 —— 避免不必要的编排开销。
对于多任务阶段,创建子工作者任务。每个工作者的描述必须是 自包含的:执行代理无法访问父级上下文、计划文档或先前的对话。描述中需包含:
目标和交付成果
相关的文件路径和模块名称
工作者成功所需的任何约束、约定或上下文
eve job create
--project $EVE_PROJECT_ID
--parent $PHASE_JOB_ID
--description "任务:<目标>。交付成果:<结果>。文件:<路径>。上下文:<工作者需要的任何信息>"
--phase ready
使阶段任务等待其子任务:
eve job dep add $PHASE_JOB_ID $TASK_JOB_ID --type waits_for
blocks 类型。工作者接收任务作业并独立执行。每个工作者:
eve job update $TASK_JOB_ID --phase active
# 执行工作
eve job submit $TASK_JOB_ID --summary "已完成 <交付成果>"
如果无需评审:
eve job close $TASK_JOB_ID --reason "完成"
编排器(根任务或阶段任务)在创建其子任务并建立依赖关系后,应返回一个等待信号。这将在子任务并行执行时释放编排器的资源:
{
"eve": {
"status": "waiting",
"summary": "已生成子任务并添加了 waits_for 关系"
}
}
编排器应保持轻量级:
--type 指定 issue_type)。bd dep add -> eve job dep add <父级> <子级> --type waits_forbd ready/blocked -> eve job dep list <id> + eve job list --phase ...如果任务需要在共享分支上进行代码更改:
eve job create \
--project $EVE_PROJECT_ID \
--description "任务:<目标>" \
--git-ref main \
--git-ref-policy explicit \
--git-branch feature/<名称> \
--git-create-branch if_missing \
--git-commit required \
--git-push on_success
保持跨任务的 Git 控制一致,以便所有更改都合并到一个 PR 中。
每周安装次数
152
仓库
首次出现
2026年2月8日
安全审计
安装于
gemini-cli152
codex152
claude-code150
pi54
opencode35
github-copilot35
If a root job already exists, use it as the orchestrator. The root epic never executes implementation work — it creates phase jobs, wires up dependencies, and waits.
Create one child job per plan phase. Each phase job acts as a sub-orchestrator: it breaks its scope into task jobs and coordinates them.
eve job create \
--project $EVE_PROJECT_ID \
--parent $EVE_JOB_ID \
--description "Phase: <name>. Deliverable: <artifact/result>" \
--phase ready
Add dependencies so the parent waits on each phase:
eve job dep add $EVE_JOB_ID $PHASE_JOB_ID --type waits_for
Split each phase into 2-6 atomic tasks with clear deliverables.
If a phase has only one task, execute it directly in the phase job rather than creating a child — avoid unnecessary orchestration overhead.
For multi-task phases, create child worker jobs. Each worker description must be self-contained : the executing agent has no access to the parent's context, the plan document, or prior conversation. Include in the description:
The objective and deliverable
Relevant file paths and module names
Any constraints, conventions, or context the worker needs to succeed
eve job create
--project $EVE_PROJECT_ID
--parent $PHASE_JOB_ID
--description "Task: <objective>. Deliverable: <result>. Files: <paths>. Context: <anything the worker needs>"
--phase ready
Make the phase wait on its tasks:
eve job dep add $PHASE_JOB_ID $TASK_JOB_ID --type waits_for
blocks only for true sequencing requirements.Workers pick up task jobs and execute them independently. Each worker:
eve job update $TASK_JOB_ID --phase active
# do the work
eve job submit $TASK_JOB_ID --summary "Completed <deliverable>"
If no review is required:
eve job close $TASK_JOB_ID --reason "Done"
After an orchestrator (root or phase) creates its child jobs and wires dependencies, it should return a waiting signal. This frees the orchestrator's resources while children execute in parallel:
{
"eve": {
"status": "waiting",
"summary": "Spawned child jobs and added waits_for relations"
}
}
Orchestrators should stay lightweight:
--type if available).bd dep add -> eve job dep add <parent> <child> --type waits_forbd ready/blocked -> eve job dep list <id> + eve job list --phase ...If tasks require code changes on a shared branch:
eve job create \
--project $EVE_PROJECT_ID \
--description "Task: <objective>" \
--git-ref main \
--git-ref-policy explicit \
--git-branch feature/<name> \
--git-create-branch if_missing \
--git-commit required \
--git-push on_success
Keep git controls consistent across tasks so all changes land in one PR.
Weekly Installs
152
Repository
First Seen
Feb 8, 2026
Security Audits
Installed on
gemini-cli152
codex152
claude-code150
pi54
opencode35
github-copilot35
Azure Data Explorer (Kusto) 查询技能:KQL数据分析、日志遥测与时间序列处理
125,100 周安装
Autoresearch:自主实验循环工具,自动化代码优化与性能改进
827 周安装
zeroize-audit 安全审计工具:检测敏感数据归零漏洞与编译器优化风险
811 周安装
App Store Connect CLI 崩溃排查工具:快速分析 TestFlight 崩溃报告与性能诊断
839 周安装
Android设备自动化工具Midscene使用指南 - 基于AI的移动端自动化测试与操作
835 周安装
.NET并发编程模式指南:async/await、Channels、Akka.NET选择决策树
820 周安装
fp-check误报检查工具:验证漏洞真伪,提升安全分析准确性
853 周安装