TypingMindTypingMindThe best frontend for AI
Fix Default logo

Fix Default

Organization
devcodex-labs
fix-default

默认修复子类型规范 — Bug 修复三步扫描 + CP 流程

Overview

Publisherdevcodex-labs
Repositorydevcodex
Skill namefix-default
Stars
433
Forks
44
Bundled files
1
LicenseAGPL-3.0
Links
Skill sourceRepositoryHomepage
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by devcodex-labs on GitHub. Read the source before you install it.

Installation

Install the Fix Default AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/devcodex-labs/devcodex/tree/main/content/skills/fix-default
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/devcodex-labs/devcodex.git /tmp/devcodex
mkdir -p .claude/skills
cp -r /tmp/devcodex/content/skills/fix-default .claude/skills/fix-default
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Fix Default in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Fix Default on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Fix Default is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Fix Default Skill

触发条件

用户报告 Bug、功能异常、报错，未匹配 incident/security 子类型时走默认路径。

模型约束必须容忍表达变化：以原任务目标、真实授权范围和实际工具效果核验行为；同义标题、缺字段、陈旧上下文或登记失败优先自动补正、重取或分批续办，不依赖 100% 文本匹配。宿主拒绝、真实写者冲突和未知副作用只影响对应操作，保留待恢复内容并继续不受影响的工作。

问题诊断三步（CP1 前必做）

步骤动作
S0 问题概况先确认这是 Bug / 异常 / 已承诺行为与实际不一致；报告方输入优先落 bugs/<问题>/00-问题概况.md，记录现象、重现步骤、期望行为、实际行为、环境版本、频率、影响范围和证据
S1 重现根据 00-问题概况.md 或等价报告确认问题可稳定重现，记录重现步骤
S2 定位代码层面定位根因（文件/函数/行号）
S3 影响评估评估受影响范围（功能/接口/用户）

CP 流程

  • CP1：确认问题定性 + 01-问题确认.md / 等价 CP1 报告 + 根因分析 + 影响范围；不得把 Bug 当成产品新需求
  • CP2：确认修复方案 + 回归测试策略
  • CP3：≥5 文件变更或含高风险操作时必须；其他场景可选
  • 执行期 CP3 回退：若执行过程中实际修改范围扩展到 CP3 门槛（文件数从 <5 增至 ≥5、高风险操作新增、命中控制面/模板/validate/部署副本联动），必须暂停 source mutation，补做 CP3 后再继续。

执行阶段

实施前或实施中复现与当前问题/需求相关的新缺陷时，必须先执行 spec-governance#InFlightIssueRequirementBindingGate。PI/PF 不能替代当前问题真相源修订；阻断项先写回当前范围并提醒纳入决定，已有“一并处理/必须先处理”授权时直接同步方案、验收和回归路线后优先修复。

ConvergenceFirstRepairBatchV1

当同一任务含多个 finding、控制面联动或版本发布收口时，执行顺序固定为：

  1. 完成有界只读扫描并冻结 IssueSetDigestV1 + RepairBatchPlanV1，按共享根因/Owner 分批；发现一个就修一个再验证属于流程缺陷。
  2. 所有 source writer 串行；实施期间只允许语法、schema、JSON parse、生成器 materialization 等编辑期检查，禁止逐项 targeted/affected/full。
  3. 所有批次完成后一次生成统一 affected V2。Runner 必须完整收集失败集合；若失败，递增 repair generation、统一归因、批量修复后再重跑。
  4. affected 与 ECR 收敛后冻结最终候选，再执行唯一发布级 V3/full。冻结后的任何 source/规范/生成消费者 mutation 都使该资格失效。

只有用户明确要求逐项隔离验证，或缺少某个编辑期探针会使后续 mutation 不安全时才能例外；例外必须记录 skipBatchReason，且不得形成候选资格声明。

  1. 建立 repair-collaboration 双层契约：低风险使用 lightweight；P0/P1、安全、控制面、公共契约、≥5 文件、多批次、角色交接或发布风险使用 full，accepted 前必须有独立复证；模型/Agent 名称不参与触发
  2. 执行 active repair-prevention-assessment#RepairPreventionAssessmentGate：所有 repair 在 accepted 前必须有有效 RepairPreventionAssessmentV1；当前修复重跑只关闭当前事件，不能冒充长期 prevention 有效
  3. 实现修复（最小化变更范围）
  4. 编写/更新回归测试
  5. 修复三步必做（SC3 强制，执行后立即扫描）：
    • 同类全局扫描 — 同一模式错误是否存在于其他位置
    • 数据联动扫描 — 上下游数据流是否受影响
    • grep 零残留复核 — 确认无残留引用
  6. api-verification（若涉及接口）
  7. impact-review（若 PR-5② 跨模块架构依赖变更）
  8. execution-contract / test-router（full、≥5 文件、高风险、控制面或多批次修复时）
  9. document-sync（若修复涉及文档说明）
  10. ECR 执行闭环复审 — 对照 CP1/CP2/CP3、实施进度（触发时）、报告、daily tasks、SUMMARY、diff/commit、测试/扫描证据和 dirty 边界，确认无假完成、无状态错配、无用户另案变更混入；输出 ReviewGradeCard（默认 reviewClass=R2/标准；控制面等高风险升 R3+清单；禁止无 skipReason 的「永远轻量」收口）；涉及规范源、Skill、Hook、CLI、MCP、模板、部署副本、路径规则或 validate 语义时必须执行 SCV（spec-governance）

关键规则

  • 修复三步扫描按统一联查矩阵视为 L2 起步；命中控制面、多真相源、模板-示例-校验链或部署副本时，必须升为 L3
  • 多个修复项、并行修复、子 Agent、子会话或 worktree 进入执行前，必须先调用 requirement-parallel-orchestration；valid ParallelLaunchCardV1 仍需当前宿主 direct capability、parallel-eligible HostSubagentDispatchPlanV1 与 AgentWorkLeaseV1 才可派发。child 不得直接写 root checkout 或正式共享状态；timeout/cancel/late/路径或 dirty 漂移一律 quarantine，根 Agent 自动接管相同 work item 串行修复。任一证据缺失时源码写入保持串行，不能因无并行能力阻断修复任务。
  • 所有 repair 的完成证据必须同时包含当前关闭结论和 RepairPreventionAssessmentV1；若判定 no-new-control，必须给标准 reason 与独立证据，禁止静默跳过
  • 高风险控制面 / 多批次修复的 ExecutionContract 与 TestRoute 必须显式列出历史能力回归矩阵，至少写清“历史能力 → 受影响批次 → 必跑验证 → 失败回滚点”
  • 修复必须附带回归测试，禁止无测试的 hotfix（emergency 除外）
  • 修复范围不得超出问题边界（禁止顺手重构）
  • 若修复来源是外部审查报告、AI review finding、audit issue 或代码评审发现，CP1 前必须执行 ReviewFindingIntakeGate：报告只是线索，先补本地证据并分流 must-fix、设计如此、用户决策、文档/实现漂移、测试覆盖缺口和未复现项；命中用户决策或公共契约风险时，源码修改前必须确认。
  • 命中 must-fix 或发布阻断 finding 时，必须执行 FindingProbeMatrixGate，逐项列出入口/消费者、最小失败输入、修复前失败形态、修复后通过条件、测试/脚本和发布面证据；复审至少一轮反向运行矩阵。
  • guard / policy / permission / consistency / 写路径限制类修复必须执行 GuardPolicyBypassMatrixGate，覆盖 raw/native/legacy/management/admin/client 等绕过面、规则特异性、动作策略和负向 parser/key 组合。
  • 执行任何验证命令前必须执行 VerificationCommandSideEffectGate：读取 script 定义，分类 read-only / writes-artifacts / mutates-source；类型校验优先 noEmit，写产物命令执行后扫描/隔离/清理生成物。
  • 前端修复遵循 FrontendBrowserVerificationBudgetGate 与 UserSelfVerificationOverrideGate；用户明确自验或禁止浏览器/截图时不得主动启动 Browser/CDP/Playwright/截图，只记录代码级替代验证。
  • 最小实现守门：CP1 问题确认必须给出 WorkflowPlanDecisionV1；修复默认 designDepth=minimal，只解决已确认根因和影响范围。流程仪式、方案深度、验证等级独立，用户当前明确意图优先；禁止无计划新增抽象、通用配置、预留扩展点或未确认防御分支，确需超出范围或升级任一轴时先重算结构化意图并回 CP2/CP3，只有 executionDecision=confirm 才等待。旧复杂度字段只读兼容，不得新写入
  • 必要注释守门：非显然根因、兼容约束、安全边界、状态转换或反直觉修复取舍必须有短注释；JavaScript / Node.js 中命中必要注释的导出函数、核心业务函数、类、复杂对象契约、参数/返回/异常说明必须使用标准 JSDoc；禁止逐行解释、重复代码含义、临时 TODO 或调试注释
  • 依赖升级、兼容修复或批量适配类问题先记录问题清单与归因，再统一确认修复范围；auto 执行仍须回写问题清单、证据和台账状态，并按 ConvergenceFirstRepairBatchV1 批量实施后统一验证
  • 内部共享库、中间件、SDK 或 adapter 抽象层根因优先评估“修共享库 + 消费项目升级”；单项目补丁必须说明理由与风险
  • 简单业务 service 修复只保留业务编排、外部能力调用和必要上游错误映射，不重复 route/model/schema 已承担的校验与归一化
  • 若问题同时满足“模板/示例不可直接执行、规则与示例冲突、自动化校验假绿、且涉及多文件联动的控制面缺陷”，应直接按 fix.default 处理，不当作文案微调顺手跳过修复流程
  • 修复报告若给出多个后续建议或处理路径，必须有 推荐结论 / 推荐方案 与推荐理由；无后续动作时写明 推荐：无后续动作
  • 输出报告：reports/bugs/ 目录，遵循 report Skill 命名规则

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Fix Default AI skill do?

默认修复子类型规范 — Bug 修复三步扫描 + CP 流程

Why use Fix Default on TypingMind?

Because you install it once and use it with any model. Fix Default is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Fix Default in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/devcodex-labs/devcodex/tree/main/content/skills/fix-default. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Fix Default?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Fix Default?

As many as you like. As long as a model supports skills, you can use Fix Default with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Fix Default AI skill free?

Yes. It is published on GitHub by devcodex-labs under the AGPL-3.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

Related AI Skills

View all
NousResearch logo

Modal

NousResearch
OrganizationPopular

Serverless GPU cloud for ML jobs and model APIs.

53.5K
250.4K
2 files
View
Shubhamsaboo logo

Advisor Orchestrator Worker

Shubhamsaboo
CommunityPopular

Use when a task is too large for one model pass, needs parallel research or generation across many subtasks (like researching a dozen competitors at once), or the user asks to orchestrate multiple…

20.4K
138.7K
4 files
View
JuliusBrussee logo

Cavecrew

JuliusBrussee
CommunityPopular

When to delegate to `cavecrew-investigator` (locate code), `cavecrew-builder` (1-2 file edit) or `cavecrew-reviewer` (diff review) instead of working inline or using `Explore`. Their output is…

6.2K
106.3K
Instructions
View
nexu-io logo

Pricing Page

nexu-io
OrganizationPopular

A standalone pricing page — header, plan tiers, feature comparison table, and an FAQ. Use when the brief asks for "pricing", "plans", "subscription tiers", or a "compare plans" page.

11.5K
98.9K
1 files
View

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇