Install
$ agentstack add skill-choshimwy-agentdevelopmentskills-workflow-orchestration ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
跨平台工作流编排
Purpose
Provide platform-neutral task classification, checkpoints, role boundaries, review independence, failure loops, and delivery reporting. This Skill never selects Xcode, Gradle, browser, backend, or device commands.
When to Use
- 任务需要跨文件实施、验证、审查或交付汇总。
- 平台包需要复用统一 checkpoint、角色或失败回环合同。
- 用户明确要求 subAgent、多 Agent、并行协作或全局任务推进。
When Not to Use
- 平台 API、构建、设备、签名或运行时诊断的具体执行。
- 单一专项 Skill 已能完整处理的纯查询。
- 用通用编排替代平台实现、验证或权限审批。
Agent Rules
analyze模式只读,输出目标、范围、风险和候选验证,不得写文件或触发执行型工具。orchestrate模式可以协调实施,但实际副作用由已解析平台/discipline Capability 承担。report模式只汇总已有证据,不新增实现、验证或审查结论。- 修复/实现任务首次写入前必须完成
CP0 Intent Lock简短计划。 - 实现链路必须是:实施 → 定向验证或
no_test_reason→ 独立 reviewer subAgent。 - reviewer 不可用或存在阻塞问题时不得宣告完成。
- 多 Session Worktree 的 Registry、Stacked 依赖与状态转换由本 Skill 管理;Git Worktree/Base/Patch 交给
session-worktree。 - Session Final Gate 只能消费 clean Checkpoint Commit 集合,并复用 Adapter Request/Result、Node Attempt 与 Run Ledger 证据;Gate 后不得再生成或改写 Commit。
- 同类失败按
fail-fix-report最多回环两次;超限后报告 blocked。 - 不在本 Skill 中硬编码平台 Skill 名、workspace、scheme、device、Gradle task、浏览器或后端部署命令。
Task Classification
| 类型 | 默认档位 | | --- | --- | | doc-only / rule-only | lite | | code-small / code-medium | standard | | code-risky | full |
最低逻辑角色为 explorer + builder + reporter。实现链路额外要求独立 reviewer;pm、tester 和资料研究角色按任务证据启用。
Checkpoints
CP0 Intent Lock:目标、边界、非目标、成功标准、计划与验证路径冻结。CP1 Anchor Slice:先完成最小关键切片,再扩大实施面。CP2 Validation Baseline Freeze:冻结受影响测试、环境与证据口径。CP3 Final Gate:验证有效且独立 review 无阻塞。
详见 references/checkpoint-contract.md 与 references/handoff-loop.md。 多 Worktree Session 详见 references/session-gates.md。
Inputs
{
"goal": "task objective",
"context": [],
"constraints": [],
"success_criteria": [],
"selected_platforms": [],
"mode": "analyze | orchestrate | report"
}
Outputs
{
"status": "completed | partial | blocked",
"task_type": "doc-only | rule-only | code-small | code-medium | code-risky",
"orchestration_level": "lite | standard | full",
"checkpoint_status": {},
"roles_used": [],
"validation": {},
"review": {},
"acceptance_matrix": [],
"known_risks": [],
"next_action": "none | blocked | needs_user_input | needs_verification"
}
Exit Conditions
- 需求与改动范围可追溯。
- 必要验证发生在最后一次相关修改之后,或明确记录
no_test_reason。 - 实现任务已由未参与实现的 reviewer 审查,且
阻塞问题:无。 - 平台专属执行证据由对应平台 Capability 提供,而非本 Skill 伪造。
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: ChoshimWy
- Source: ChoshimWy/AgentDevelopmentSkills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.