Jeden Skill in Manus ausführen
mit einem Klick

Jeden Skill in Manus mit einem Klick ausführen

execution-guard

Sterne3

Forks2

Aktualisiert25. März 2026 um 21:43

Use for any coding task: fix bug, implement feature, update file, refactor, debug, investigate failure, add tests, run validators, review changes, or verify a repo. Enforces clarify → explore → route → TDD → verify → review for a non-programmer user.

Installation

Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.

In Manus ausführen

Quelle

ryan-mt

ryan-mt/claude-backup

GitHub-Repository öffnen Creator-Repositorys ansehen

Download

In Manus ausführen

Verwandte BerufeSOC

Basierend auf der SOC-Berufsklassifikation

SoftwareentwicklerInformatik- und Mathematikberufe·SOC 15-1252

SKILL.md

readonly

name	execution-guard
description	Use for any coding task: fix bug, implement feature, update file, refactor, debug, investigate failure, add tests, run validators, review changes, or verify a repo. Enforces clarify → explore → route → TDD → verify → review for a non-programmer user.

Execution Guard

Apply this workflow on coding tasks.

1. Clarify only what matters

If desired behavior, scope, or acceptance criteria are unclear, ask concise product questions before editing
Do not ask the user for technical implementation choices unless unavoidable
Never guess missing behavior just to keep moving

2. Explore before editing

Read the relevant files first
Find existing patterns before introducing new ones
Discover the repo's real validator commands from project config before coding

3. Route work by cost and skill

repo-scout (haiku): search, inventory, conventions, validator discovery
planner (sonnet): multi-file planning, risks, open questions
code-reviewer (sonnet): post-edit review
validator (sonnet): build, typecheck, lint, tests, coverage
security-reviewer (opus): before commits or security-sensitive changes
Keep Opus as orchestrator for ambiguity, architecture, and security-sensitive decisions

4. TDD is the default

RED: write or update a failing test first when behavior changes
GREEN: make the smallest change that passes
REFACTOR: improve while keeping tests green
If the repo has no test harness, state that clearly and use the closest available safety net

5. Verification is evidence, not vibes

Run the repo's actual validators before claiming completion
Record exact command, exit code, and key output for each check
If a check is skipped, unavailable, or flaky, say so explicitly
Never say "verified" because the code looks right

6. Review before handoff

Re-read edited files
Run code-reviewer after validators pass
Before any commit, run security-reviewer

7. Honesty rules

Separate facts from assumptions
Trust filesystem and command output over memory
If something is still uncertain or blocked, say that directly

Mehr aus diesem Repository

gleiches Repository

anti-ai-design

ryan-mt/claude-backup

Enforce distinctive, non-AI-looking design for any UI task across ALL stacks. Auto-triggers on: build UI, page, component, layout, web app, design, style, CSS, animation, theme, landing page, dashboard, form, navigation, frontend, website, interface, responsive, mobile, header, footer, sidebar, hero, card, button, modal, Tauri, Electron, React Native, Flutter, SwiftUI, desktop app, mobile app, native app.

2026-03-253

deep-research

ryan-mt/claude-backup

Conducts enterprise-grade research with multi-source synthesis, citation tracking, and verification. Produces citation-backed reports through a structured pipeline with source credibility scoring. Triggers on "deep research", "comprehensive analysis", "research report", "compare X vs Y", "analyze trends", or "state of the art". Not for simple lookups, debugging, or questions answerable with 1-2 searches.

2026-03-253

planning-with-files

ryan-mt/claude-backup

Transforms workflow to use Manus-style persistent markdown files for planning, progress tracking, and knowledge storage. Use when starting complex tasks, multi-step projects, research tasks, or when the user mentions planning, organizing work, tracking progress, or wants structured output.

2026-03-253

seo-audit

ryan-mt/claude-backup

Full website SEO audit with parallel subagent delegation. Crawls up to 500 pages, detects business type, delegates to 7 specialists, generates health score. Use when user says "audit", "full SEO check", "analyze my site", or "website health check".

2026-03-253

seo-audit

ryan-mt/claude-backup

When the user wants to audit, review, or diagnose SEO issues on their site. Also use when the user mentions "SEO audit," "technical SEO," "why am I not ranking," "SEO issues," "on-page SEO," "meta tags review," "SEO health check," "my traffic dropped," "lost rankings," "not showing up in Google," "site isn't ranking," "Google update hit me," "page speed," "core web vitals," "crawl errors," or "indexing issues." Use this even if the user just says something vague like "my SEO is bad" or "help with SEO" — start with an audit. For building pages at scale to target keywords, see programmatic-seo. For adding structured data, see schema-markup. For AI search optimization, see ai-seo.

2026-03-253

seo-competitor-pages

ryan-mt/claude-backup

Generate SEO-optimized competitor comparison and alternatives pages. Covers "X vs Y" layouts, "alternatives to X" pages, feature matrices, schema markup, and conversion optimization. Use when user says "comparison page", "vs page", "alternatives page", "competitor comparison", or "X vs Y".

2026-03-253

name	execution-guard
description	Use for any coding task: fix bug, implement feature, update file, refactor, debug, investigate failure, add tests, run validators, review changes, or verify a repo. Enforces clarify → explore → route → TDD → verify → review for a non-programmer user.

Execution Guard

Apply this workflow on coding tasks.

1. Clarify only what matters

If desired behavior, scope, or acceptance criteria are unclear, ask concise product questions before editing
Do not ask the user for technical implementation choices unless unavoidable
Never guess missing behavior just to keep moving

2. Explore before editing

Read the relevant files first
Find existing patterns before introducing new ones
Discover the repo's real validator commands from project config before coding

3. Route work by cost and skill

repo-scout (haiku): search, inventory, conventions, validator discovery
planner (sonnet): multi-file planning, risks, open questions
code-reviewer (sonnet): post-edit review
validator (sonnet): build, typecheck, lint, tests, coverage
security-reviewer (opus): before commits or security-sensitive changes
Keep Opus as orchestrator for ambiguity, architecture, and security-sensitive decisions

4. TDD is the default

RED: write or update a failing test first when behavior changes
GREEN: make the smallest change that passes
REFACTOR: improve while keeping tests green
If the repo has no test harness, state that clearly and use the closest available safety net

5. Verification is evidence, not vibes

Run the repo's actual validators before claiming completion
Record exact command, exit code, and key output for each check
If a check is skipped, unavailable, or flaky, say so explicitly
Never say "verified" because the code looks right

6. Review before handoff

Re-read edited files
Run code-reviewer after validators pass
Before any commit, run security-reviewer

7. Honesty rules

Separate facts from assumptions
Trust filesystem and command output over memory
If something is still uncertain or blocked, say that directly