10x-mode
Route a hard engineering task to one existing workflow. Use for a bug fix, refactor, feature, architecture or investigation, parallel work, or finishing a branch or pull request, including when the user asks for 10x-mode.
- License 许可
- MIT
- 最近更新
- 2026-11-12
- Stars
- ★ 1
- Forks
- 0
01 前言 Front Matter
- name
- 10x-mode
- uses
- [ "[[plugin/10x-swe/diagnose/SKILL]]", "[[plugin/10x-swe/parallel-debugging/SKILL]]", "[[plugin/10x-swe/test-driven-development/SKILL]]", "[[plugin/10x-swe/refactor/SKILL]]", "[[plugin/10x-swe/brainstorming/SKILL]]", "[[plugin/10x-swe/improve-codebase-architecture/SKILL]]", "[[plugin/10x-swe/code-research/SKILL]]", "[[plugin/10x-swe/grill-me/SKILL]]", "[[plugin/10x-swe/grill-with-docs/SKILL]]", "[[plugin/10x-swe/deep-architecture-analysis/SKILL]]", "[[plugin/10x-swe/dispatching-parallel-agents/SKILL]]", "[[plugin/10x-swe/subagent-driven-development/SKILL]]", "[[plugin/10x-swe/task-coordination-strategies/SKILL]]", "[[plugin/10x-swe/finishing-a-development-branch/SKILL]]", "[[plugin/10x-swe/using-git-worktrees/SKILL]]", "[[plugin/10x-swe/find-skills/SKILL]]" ]
- metadata
- { "skill-graph-id": "plugin/10x-swe/10x-mode", "skill-graph-uses": "plugin/10x-swe/diagnose, plugin/10x-swe/parallel-debugging, plugin/10x-swe/test-driven-development, plugin/10x-swe/refactor, plugin/10x-swe/brainstorming, plugin/10x-swe/improve-codebase-architecture, plugin/10x-swe/code-research, plugin/10x-swe/grill-me, plugin/10x-swe/grill-with-docs, plugin/10x-swe/deep-architecture-analysis, plugin/10x-swe/dispatching-parallel-agents, plugin/10x-swe/subagent-driven-development, plugin/10x-swe/task-coordination-strategies, plugin/10x-swe/finishing-a-development-branch, plugin/10x-swe/using-git-worktrees, plugin/10x-swe/find-skills", "skill-graph-used-by": "" }
- description
- Route a hard engineering task to one existing workflow. Use for a bug fix, refactor, feature, architecture or investigation, parallel work, or finishing a branch or pull request, including when the user asks for 10x-mode.
02 资源文件 6
- 参考文档playbooks/architecture.md904 B
- 参考文档playbooks/bug-fix.md767 B
- 参考文档playbooks/feature.md747 B
- 参考文档playbooks/finish.md272 B
- 参考文档playbooks/parallel.md596 B
- 参考文档playbooks/refactor.md594 B
03 SKILL.md 正文
10x-mode
Pick one playbook, copy its steps into the todo list, and follow the workflow skills those steps name. Do not restate their procedures here.
Rules
- Restate the goal and the smallest change that meets it before editing.
- Follow the repository decision order: correctness, safety, observability, operational simplicity, maintainability, portability, performance, then elegance.
- Verify the real outcome before calling the task done. Label each claim measured, inferred, or guess in the same sentence. A prediction is a guess. Run a check you can run instead of handing it back.
- Destructive, secret, authentication-boundary, production, and publication actions need explicit authorization and a rollback. Delegation, installation, and network access need the same authorization. A skill link does not grant them.
- Preserve unrelated work. Commit, push, merge, or deploy only when that step is authorized.
- When the work depends on a domain under Domain references, read that skill's
SKILL.mdbefore editing and open only the sections it points at. If nothing listed fits, follow [[plugin/10x-swe/find-skills/SKILL]]. Discovery does not authorize installing or running an external skill.
Playbooks
Copy the matched playbook's numbered steps into the todo list before any
task-specific todos. Keep a step you do not take, and write skip: plus the
reason on that item. Then follow the playbook.
| Task | Playbook |
|---|---|
| A bug or performance regression to reproduce and fix | bug-fix opens [[plugin/10x-swe/diagnose/SKILL]], [[plugin/10x-swe/parallel-debugging/SKILL]], and [[plugin/10x-swe/test-driven-development/SKILL]] |
| A behavior-preserving structure change | refactor opens [[plugin/10x-swe/refactor/SKILL]] and [[plugin/10x-swe/test-driven-development/SKILL]] |
| New or changed behavior | feature opens [[plugin/10x-swe/brainstorming/SKILL]], [[plugin/10x-swe/test-driven-development/SKILL]], and [[plugin/10x-swe/improve-codebase-architecture/SKILL]] |
| A question, design, or investigation without an agreed code change | architecture opens [[plugin/10x-swe/code-research/SKILL]], [[plugin/10x-swe/improve-codebase-architecture/SKILL]], [[plugin/10x-swe/grill-me/SKILL]], [[plugin/10x-swe/grill-with-docs/SKILL]], and [[plugin/10x-swe/deep-architecture-analysis/SKILL]] |
| Two or more independent tasks, or a plan of independent tasks | parallel opens [[plugin/10x-swe/dispatching-parallel-agents/SKILL]], [[plugin/10x-swe/subagent-driven-development/SKILL]], and [[plugin/10x-swe/task-coordination-strategies/SKILL]] |
| Implementation is done and the branch needs a merge, PR, or cleanup decision | finish opens [[plugin/10x-swe/finishing-a-development-branch/SKILL]] and [[plugin/10x-swe/using-git-worktrees/SKILL]] |
A defect found during feature or refactor work uses bug-fix for that defect. Parallel work still follows the playbook for each task. When two playbooks fit, do the one that produces the requested artifact, and keep the other as a later todo if the request includes it.
Domain references
These skills do not auto-load. Read the file. On hosts whose skill tool skips manual-only skills, the file read is the load path.
- Python references:
python-anti-patterns,python-background-jobs,python-code-style,python-configuration,python-design-patterns,python-error-handling,python-observability,python-resilience,python-resource-management,python-type-safety - Python workflows that still auto-load:
async-python-patterns,python-performance-optimization,python-testing-patterns - Python packages:
uv-package-manager - LLM applications:
langchain-architecture,llm-evaluation - Go concurrency:
go-concurrency-patterns - Data and services:
sql-optimization-patterns,cqrs-implementation,microservices-patterns,error-handling-patterns,memory-safety-patterns - Interfaces and docs:
e2e-testing-patterns,agent-browser,chrome-devtools,prompt-engineering-patterns,architecture-decision-records,obsidian-markdown,bash-defensive-patterns
Search when the name is not in this list:
skill-search search 'the task' --limit 5
The router shape adapts pstack poteto-mode under the MIT license.
04 同仓库更多 Skills
turnstile-spin
Set up Cloudflare Turnstile end-to-end in a project. Scan the codebase, create the widget via the Cloudflare API, embed it where user requests need bot verification (form submissions, SPA actions, API endpoints, download links, comment or vote submissions, etc.), wire canonical server-side siteverify in the customer's existing backend, validate, and persist the skill. Load this when a user asks to add Turnstile, set up CAPTCHA, protect a form or endpoint from bots, or fix a Turnstile integration. Uses Cloudflare's documented Turnstile contract.
frontend-design
Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
source-insights
Prepare inspected source evidence, preserve exact citations, and publish reviewed insights into an explicit local library.
skill-graph
Discover skills and workflows through independent lexical search, relationship paths and a terminal graph.
session-search
Recall public session turns with exact provenance, bounded excerpts and independent piped exports.
newsletter
Produce a private local AI newsletter from bounded public research, with cited deep dives, HTML/Markdown output, Dagu scheduling and seven-day retention.