技能 executor-planning
📦
executor-planning
内容修订 r1 中风险 ⚙️ 外部命令📁 文件系统访问
Turn Approved Specifications into Implementation Plans
Incomplete implementation plans leave developers guessing about interfaces, dependencies, and acceptance criteria. This skill structures approved specifications into traceable tasks with explicit tests and execution gates.
支持: Claude Codex Code(CC)
使用我的智能体安装
将此请求复制给你的智能体，其中包含权威 Skill 页面和清单。
智能体请求
Review the Skillstore skill "executor-planning" from https://skillstore.io/skills/atri10-executor-planning.md and its manifest at https://skillstore.io/api/skills/atri10-executor-planning/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.你的智能体仍应展示其计划，并按安全策略请求必要确认。
Agent 可读资源
当 AI Agent、爬虫或脚本需要干净上下文，而不是读取完整页面时，请使用这些链接。
测试它
正在使用“executor-planning”。 Plan an approved feature that adds scoring and routing through documented interfaces.
预期结果:
- Draft plan: scoring and routing, with requirement coverage and exact interface contracts.
- Task sequence: define shared types, implement scoring, then integrate routing.
- Dependency map: routing consumes shared types and scoring results.
- Acceptance checks: explicit expected values, negative-input tests, and repeatable test commands.
- Handoff: complete plan regression before asking the human to select an execution mode.
正在使用“executor-planning”。 Review a draft plan whose final task contains a dependency section and omits an expected test failure.
预期结果:
- Extraction defect: move the dependency section above the first task heading.
- Test defect: name the exact expected failure before implementation exists.
- Coverage check: associate every requirement with at least one task.
- Verification status: rerun task extraction and plan linting with the companion tools before claiming the gate passed.
安全审计
中风险v1 • 2026/10/5 打开版本化报告
Most of the 221 static findings are Markdown formatting, legitimate companion references, or routine planning commands. Two temporary-file findings remain confirmed, and a semantic finding identifies overly broad Git staging. No evidence found of prompt injection or exfiltration; companion script implementations are outside the reviewed package.
1
已扫描文件
773
分析行数
2
复核项
0
已忽略误报
已确认安全问题 (1)
中
Broad Git Staging Can Include Unrelated Changes
The handoff runs 'git add docs/executor/INIT-0004-cloud-tenant-cells' before committing. Directory-wide staging can include unrelated or sensitive initiative files beyond the intended plan, index, and specification updates.
The example explicitly stages the entire initiative directory instead of the three intended document updates. Actual exposure depends on other changes in that directory.
能力复核项 (2)
这些是真实的本地能力，对此技能可能是预期行为，因此需要复核，但不会按已确认恶意行为计分。
中
Temp directory access
"$n" /tmp/brief-check.md >/dev/null || echo "TASK $n FAILS"
The example writes each brief to a predictable shared temporary pathname without showing exclusive creation. Concurrent runs or precreated links could corrupt or redirect output.
中
Temp directory access
grep -m1 '^\*\*Task:\*\*' /tmp/brief-check.md
The loop reads the same shared temporary file even after extraction failure. Stale output or another concurrent run could substitute the task identity being verified.
风险因素
⚙️ 外部命令 (50)
SKILL.md:8 SKILL.md:21 SKILL.md:31 SKILL.md:32 SKILL.md:33 SKILL.md:34 SKILL.md:35 SKILL.md:36 SKILL.md:37 SKILL.md:39 SKILL.md:40 SKILL.md:41 SKILL.md:42 SKILL.md:44-45 SKILL.md:46 SKILL.md:47 SKILL.md:49 SKILL.md:54 SKILL.md:56 SKILL.md:65 SKILL.md:66 SKILL.md:68 SKILL.md:69 SKILL.md:70 SKILL.md:75 SKILL.md:76 SKILL.md:77 SKILL.md:79-83 SKILL.md:83-85 SKILL.md:85 SKILL.md:86 SKILL.md:88-93 SKILL.md:93-95 SKILL.md:95-96 SKILL.md:96-101 SKILL.md:101 SKILL.md:113-116 SKILL.md:116-119 SKILL.md:119-121 SKILL.md:121 SKILL.md:122 SKILL.md:126-128 SKILL.md:128 SKILL.md:129-132 SKILL.md:132-167 SKILL.md:167-169 SKILL.md:169 SKILL.md:176 SKILL.md:177 SKILL.md:190-193
审计者: codex
分享与引用此报告
分享版本化评估报告、中性徽章、嵌入卡片和引用。Skillstore 只报告证据，不判断该 Skill 是否安全。
复制报告链接
https://skillstore.io/skills/atri10-executor-planning/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown 徽章
[](https://skillstore.io/skills/atri10-executor-planning?utm_source=security_passport_badge)HTML 徽章
<a href="https://skillstore.io/skills/atri10-executor-planning?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/atri10-executor-planning/security.svg" alt="Skillstore security assessment" loading="lazy"></a>嵌入卡片
<iframe src="https://skillstore.io/embed/skills/atri10-executor-planning.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>学术引用 (APA · BibTeX · CFF)
APA 引用
atri10. (2026). executor-planning security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/atri10-executor-planning/audits/1BibTeX 引用
@techreport{atri10-atri10-executor-planning-2026,
author = {atri10},
title = {executor-planning security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/atri10-executor-planning/audits/1},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "executor-planning security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
- name: "atri10"
date-released: "2026-10-05"
url: "https://skillstore.io/skills/atri10-executor-planning/audits/1"
identifiers:
- type: other
value: "skillstore:atri10-executor-planning:audit:1"
description: "Skillstore immutable audit report identifier"
Skillstore 评分
评分依据 证据置信度: 中 55
架构
85
可维护性
87
内容
65
社区
74
规范符合性
你能构建什么
Prepare a Reviewable Delivery Plan
Convert an approved feature specification into traceable tasks with explicit file ownership, interfaces, and acceptance tests.
Create Self-Contained Agent Briefs
Prepare tasks that isolated coding agents can implement without access to the surrounding conversation or complete plan.
Repair a Multi-Plan Handoff
Find missing requirement coverage, interface mismatches, dependency cycles, and incomplete test steps before implementation begins.
试试这些提示
Create a First Plan
Create a draft implementation plan from [approved specification] and [interface document]. Check prerequisites and identify missing inputs before drafting tasks.
Define Testable Tasks
Refine [draft plan] into self-contained tasks. Include requirement IDs, exact interfaces, file lists, failing tests, expected results, and refactoring steps.
Split an Approved Specification
Use [decided decomposition session] to split [approved specification] into independently testable plans. Document coverage, predecessor assumptions, execution order, and dependency maps.
Audit and Repair the Plan Set
Audit [plan set] against [specification] and [interface contracts]. Repair coverage gaps, dependency cycles, and extraction defects. Report unavailable validation tools without claiming success.
最佳实践
- Confirm specification approval, interface contracts, and the human-selected decomposition before drafting tasks.
- Make each task independently understandable with exact requirements, interfaces, tests, and dependency references.
- Verify companion tools, use private temporary files, and inspect explicitly staged changes before the execution handoff.
避免
- Inventing interfaces or architectural decisions inside an implementation plan.
- Using placeholders, implicit constraints, or references to another task instead of a complete acceptance contract.
- Starting implementation before regression clears and the human selects an execution mode.
常见问题
Which AI tools are listed as supported?
The report lists Claude, Codex, and Claude Code. The workflow still requires the companion Executor resources and project documents.
What inputs are required before planning?
Provide an approved specification, relevant architecture decisions, and a decided decomposition session. Work spanning components also requires interface documents.
Does this package include the validation scripts?
No. This package contains only the planning instructions. The referenced scripts and supporting documents belong to the companion Executor package.
Can the skill invent missing interfaces?
No. Missing interfaces require an architecture decision before planning resumes. Task signatures must match the approved interface document.
When does implementation begin?
Implementation begins after plan regression clears and the human chooses subagent or inline execution. Planning does not authorize implementation by itself.
What security precautions apply to the examples?
Replace the shared temporary pathname with a private temporary directory. Stage only reviewed files and inspect staged changes before committing.