How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

create - Prompt - OpenSmartRoute

Name: create
Author: ondrasek

Prompt file imported from ondrasek/acforge-cli (.claude/commands/issue/create.md). Copyright stays with the author.

GitHub Issue Creation - Mandatory 8-Step Quality Workflow

!git status !git branch --show-current

Create new GitHub Issue with comprehensive quality validation using extended actor/critic pattern. All issues go through the complete 8-step refinement process for maximum quality assurance.

Mandatory 8-Step Workflow Instructions

Extended Actor/Critic Pattern for Content Quality + Priority Validation:

Step 1: Draft Creation

Use Task tool to delegate to github-issues-workflow agent:

Prompt user for initial issue description and gather requirements
Create comprehensive draft with:
- Clear problem description
- Initial acceptance criteria
- Basic implementation scope
Focus on completeness over perfection
Return draft content for refinement review

Step 2: Refine Content

Use Task tool to delegate to github-issues-workflow agent:

Improve draft based on GitHub issue best practices
Enhance clarity, specificity, and actionability
Add technical details and context
Structure content for developer comprehension
Return refined content for criticism

Step 3: Criticize Refinement

Use Task tool to delegate to critic agent with content quality focus:

Content Critic Prompt: "Critically analyze this issue content for quality, clarity, and actionability. Challenge generic language and identify improvement opportunities:

CONTENT QUALITY CRITERIA:

Anti-Generic Detection: Flag templated language, empty phrases, checklist padding
Specificity Challenge: Demand concrete examples, specific behaviors, measurable outcomes
Actionability Test: Verify a developer can understand exactly what to build
Value Validation: Confirm clear user/business value articulation
Scope Clarity: Ensure implementation boundaries are well-defined
Technical Depth: Assess if sufficient technical context is provided

CRITICISM ACTIONS:

Identify vague or generic content requiring specificity
Challenge unclear implementation scope or acceptance criteria
Point out missing technical details or context
Suggest specific improvements for clarity and actionability
Flag any 'stupid checklists' or meaningless template content

CONSTRUCTIVE FOCUS: Provide specific, actionable feedback for content improvement. Challenge the author to be more precise and valuable."

Step 4: Adjust Based on Criticism

Use Task tool to delegate to github-issues-workflow agent:

Address specific feedback from content critic
Eliminate vague language and generic phrases
Add concrete examples and technical specifics
Clarify implementation scope and acceptance criteria
Improve overall issue quality and actionability
Return adjusted content for labeling

Step 5: Label Assignment

Use Task tool to delegate to github-issues-workflow agent:

Discover repository labels via gh label list
Apply appropriate type labels (feat, bug, docs, refactor, etc.)
Add workflow labels (dependencies, breaking change, etc.)
Include quality labels (needs refinement, risky, etc.) if applicable
Avoid conflicting label combinations
Return content with proposed labels for validation

Step 6: Prioritize with BINARY CONFIDENCE SYSTEM

Use Task tool to delegate to github-issues-workflow agent:

Apply existing BINARY CONFIDENCE SYSTEM with strict 6-criteria assessment
Analyze for priority indicators and conflicting evidence
Determine confidence level (Confident/Uncertain)
Assign priority label only if confident (critical/high priority)
Default to medium priority (no label) if uncertain
Return content with priority assignment and confidence justification

Step 7: Criticize Labels

Use Task tool to delegate to critic agent with label validation focus:

Label Critic Prompt: "Validate the label assignments for appropriateness and accuracy:

LABEL VALIDATION CRITERIA:

Type Label Accuracy: Verify feat/bug/docs/refactor matches actual content
Conflict Detection: Identify incompatible label combinations
Completeness Check: Ensure required labels are present
Repository Consistency: Validate against established labeling patterns
Workflow Labels: Assess dependencies, breaking change, risky labels appropriateness

VALIDATION ACTIONS:

Challenge inappropriate or conflicting labels
Suggest missing required labels
Validate label consistency with issue content
Recommend label adjustments if needed

SKEPTICAL FOCUS: Question label assignments that don't clearly match issue content."

Step 8: Criticize Priority

Use Task tool to delegate to critic agent with priority validation focus:

Priority Critic Prompt: "Analyze the priority classification using the BINARY CONFIDENCE SYSTEM validation:

VALIDATION CRITERIA:

Verify 'Confident' claims only if ALL 6 strict criteria were genuinely met
Challenge confidence claims that lack sufficient evidence
Look for overlooked conflicting indicators (enhancement, nice to have, future, optional)
Validate precedent claims by checking cited similar issues
Assess whether reasoning is truly falsifiable with specific evidence
Question if 3+ keywords/indicators were actually present and relevant

VALIDATION ACTIONS:

If 'Confident' claim is unjustified: Remove priority label (default to medium)
If 'Uncertain' but strong evidence exists: Consider adding priority label
Provide Priority Validation assessment
Adjust labels only when confidence assessment was incorrect

SKEPTICAL FOCUS: Be particularly skeptical of 'Confident' claims - prevent priority inflation."

Step 9: Create Final Issue

Use Task tool to delegate to github-issues-workflow agent:

Create GitHub issue with refined content and validated labels/priority
Add comprehensive analysis comments:
- Content Quality Assessment with refinement summary
- Label Assignment Rationale
- Priority Analysis with confidence justification
- Validation Results from critic reviews
Return issue number, URL, and quality assessment summary

Final Confirmation

Provide comprehensive issue details with:

Direct GitHub issue link
Final content quality assessment
Label assignment summary
Priority classification with confidence level
Summary of all agent assessments and refinements

Install create and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

# GitHub Issue Creation - Mandatory 8-Step Quality Workflow !`git status` !`git branch --show-current` Create new GitHub Issue with comprehensive quality validation using extended actor/critic pattern. All issues go through the complete 8-step refinement process for maximum quality assurance. ## Mandatory 8-Step Workflow Instructions **Extended Actor/Critic Pattern for Content Quality + Priority Validation**: ### **Step 1: Draft Creation** Use Task tool to delegate to github-issues-workflow agent: - Prompt user for initial issue description and gather requirements - Create comprehensive draft with: - Clear problem description - Initial acceptance criteria - Basic implementation scope - Focus on completeness over perfection - Return draft content for refinement review ### **Step 2: Refine Content** Use Task tool to delegate to github-issues-workflow agent: - Improve draft based on GitHub issue best practices - Enhance clarity, specificity, and actionability - Add technical details and context - Structure content for developer comprehension - Return refined content for criticism ### **Step 3: Criticize Refinement** Use Task tool to delegate to critic agent with content quality focus: **Content Critic Prompt**: "Critically analyze this issue content for quality, clarity, and actionability. Challenge generic language and identify improvement opportunities: **CONTENT QUALITY CRITERIA**: - **Anti-Generic Detection**: Flag templated language, empty phrases, checklist padding - **Specificity Challenge**: Demand concrete examples, specific behaviors, measurable outcomes - **Actionability Test**: Verify a developer can understand exactly what to build - **Value Validation**: Confirm clear user/business value articulation - **Scope Clarity**: Ensure implementation boundaries are well-defined - **Technical Depth**: Assess if sufficient technical context is provided **CRITICISM ACTIONS**: - Identify vague or generic content requiring specificity - Challenge unclear implementation scope or acceptance criteria - Point out missing technical details or context - Suggest specific improvements for clarity and actionability - Flag any 'stupid checklists' or meaningless template content **CONSTRUCTIVE FOCUS**: Provide specific, actionable feedback for content improvement. Challenge the author to be more precise and valuable." ### **Step 4: Adjust Based on Criticism** Use Task tool to delegate to github-issues-workflow agent: - Address specific feedback from content critic - Eliminate vague language and generic phrases - Add concrete examples and technical specifics - Clarify implementation scope and acceptance criteria - Improve overall issue quality and actionability - Return adjusted content for labeling ### **Step 5: Label Assignment** Use Task tool to delegate to github-issues-workflow agent: - Discover repository labels via `gh label list` - Apply appropriate type labels (feat, bug, docs, refactor, etc.) - Add workflow labels (dependencies, breaking change, etc.) - Include quality labels (needs refinement, risky, etc.) if applicable - Avoid conflicting label combinations - Return content with proposed labels for validation ### **Step 6: Prioritize with BINARY CONFIDENCE SYSTEM** Use Task tool to delegate to github-issues-workflow agent: - Apply existing BINARY CONFIDENCE SYSTEM with strict 6-criteria assessment - Analyze for priority indicators and conflicting evidence - Determine confidence level (Confident/Uncertain) - Assign priority label only if confident (critical/high priority) - Default to medium priority (no label) if uncertain - Return content with priority assignment and confidence justification ### **Step 7: Criticize Labels** Use Task tool to delegate to critic agent with label validation focus: **Label Critic Prompt**: "Validate the label assignments for appropriateness and accuracy: **LABEL VALIDATION CRITERIA**: - **Type Label Accuracy**: Verify feat/bug/docs/refactor matches actual content - **Conflict Detection**: Identify incompatible label combinations - **Completeness Check**: Ensure required labels are present - **Repository Consistency**: Validate against established labeling patterns - **Workflow Labels**: Assess dependencies, breaking change, risky labels appropriateness **VALIDATION ACTIONS**: - Challenge inappropriate or conflicting labels - Suggest missing required labels - Validate label consistency with issue content - Recommend label adjustments if needed **SKEPTICAL FOCUS**: Question label assignments that don't clearly match issue content." ### **Step 8: Criticize Priority** Use Task tool to delegate to critic agent with priority validation focus: **Priority Critic Prompt**: "Analyze the priority classification using the BINARY CONFIDENCE SYSTEM validation: **VALIDATION CRITERIA**: - Verify 'Confident' claims only if ALL 6 strict criteria were genuinely met - Challenge confidence claims that lack sufficient evidence - Look for overlooked conflicting indicators (enhancement, nice to have, future, optional) - Validate precedent claims by checking cited similar issues - Assess whether reasoning is truly falsifiable with specific evidence - Question if 3+ keywords/indicators were actually present and relevant **VALIDATION ACTIONS**: - If 'Confident' claim is unjustified: Remove priority label (default to medium) - If 'Uncertain' but strong evidence exists: Consider adding priority label - Provide Priority Validation assessment - Adjust labels only when confidence assessment was incorrect **SKEPTICAL FOCUS**: Be particularly skeptical of 'Confident' claims - prevent priority inflation." ### **Step 9: Create Final Issue** Use Task tool to delegate to github-issues-workflow agent: - Create GitHub issue with refined content and validated labels/priority - Add comprehensive analysis comments: - Content Quality Assessment with refinement summary - Label Assignment Rationale - Priority Analysis with confidence justification - Validation Results from critic reviews - Return issue number, URL, and quality assessment summary ## Final Confirmation Provide comprehensive issue details with: - Direct GitHub issue link - Final content quality assessment - Label assignment summary - Priority classification with confidence level - Summary of all agent assessments and refinements

{ "prompt": "# GitHub Issue Creation - Mandatory 8-Step Quality Workflow\n\n!`git status`\n!`git branch --show-current`\n\nCreate new GitHub Issue with comprehensive quality validation using extended actor/critic pattern. All issues go through the complete 8-step refinement process for maximum quality assurance.\n\n## Mandatory 8-Step Workflow Instructions\n\n**Extended Actor/Critic Pattern for Content Quality + Priority Validation**:\n\n### **Step 1: Draft Creation**\nUse Task tool to delegate to github-issues-workflow agent:\n- Prompt user for initial issue description and gather requirements\n- Create comprehensive draft with:\n - Clear problem description\n - Initial acceptance criteria\n - Basic implementation scope\n- Focus on completeness over perfection\n- Return draft content for refinement review\n\n### **Step 2: Refine Content** \nUse Task tool to delegate to github-issues-workflow agent:\n- Improve draft based on GitHub issue best practices\n- Enhance clarity, specificity, and actionability\n- Add technical details and context\n- Structure content for developer comprehension\n- Return refined content for criticism\n\n### **Step 3: Criticize Refinement**\nUse Task tool to delegate to critic agent with content quality focus:\n\n**Content Critic Prompt**: \"Critically analyze this issue content for quality, clarity, and actionability. Challenge generic language and identify improvement opportunities:\n\n**CONTENT QUALITY CRITERIA**:\n- **Anti-Generic Detection**: Flag templated language, empty phrases, checklist padding\n- **Specificity Challenge**: Demand concrete examples, specific behaviors, measurable outcomes \n- **Actionability Test**: Verify a developer can understand exactly what to build\n- **Value Validation**: Confirm clear user/business value articulation\n- **Scope Clarity**: Ensure implementation boundaries are well-defined\n- **Technical Depth**: Assess if sufficient technical context is provided\n\n**CRITICISM ACTIONS**:\n- Identify vague or generic content requiring specificity\n- Challenge unclear implementation scope or acceptance criteria \n- Point out missing technical details or context\n- Suggest specific improvements for clarity and actionability\n- Flag any 'stupid checklists' or meaningless template content\n\n**CONSTRUCTIVE FOCUS**: Provide specific, actionable feedback for content improvement. Challenge the author to be more precise and valuable.\"\n\n### **Step 4: Adjust Based on Criticism**\nUse Task tool to delegate to github-issues-workflow agent:\n- Address specific feedback from content critic\n- Eliminate vague language and generic phrases\n- Add concrete examples and technical specifics\n- Clarify implementation scope and acceptance criteria\n- Improve overall issue quality and actionability\n- Return adjusted content for labeling\n\n### **Step 5: Label Assignment**\nUse Task tool to delegate to github-issues-workflow agent:\n- Discover repository labels via `gh label list`\n- Apply appropriate type labels (feat, bug, docs, refactor, etc.)\n- Add workflow labels (dependencies, breaking change, etc.)\n- Include quality labels (needs refinement, risky, etc.) if applicable\n- Avoid conflicting label combinations\n- Return content with proposed labels for validation\n\n### **Step 6: Prioritize with BINARY CONFIDENCE SYSTEM**\nUse Task tool to delegate to github-issues-workflow agent:\n- Apply existing BINARY CONFIDENCE SYSTEM with strict 6-criteria assessment\n- Analyze for priority indicators and conflicting evidence\n- Determine confidence level (Confident/Uncertain)\n- Assign priority label only if confident (critical/high priority)\n- Default to medium priority (no label) if uncertain\n- Return content with priority assignment and confidence justification\n\n### **Step 7: Criticize Labels**\nUse Task tool to delegate to critic agent with label validation focus:\n\n**Label Critic Prompt**: \"Validate the label assignments for appropriateness and accuracy:\n\n**LABEL VALIDATION CRITERIA**:\n- **Type Label Accuracy**: Verify feat/bug/docs/refactor matches actual content\n- **Conflict Detection**: Identify incompatible label combinations\n- **Completeness Check**: Ensure required labels are present\n- **Repository Consistency**: Validate against established labeling patterns\n- **Workflow Labels**: Assess dependencies, breaking change, risky labels appropriateness\n\n**VALIDATION ACTIONS**:\n- Challenge inappropriate or conflicting labels\n- Suggest missing required labels\n- Validate label consistency with issue content\n- Recommend label adjustments if needed\n\n**SKEPTICAL FOCUS**: Question label assignments that don't clearly match issue content.\"\n\n### **Step 8: Criticize Priority**\nUse Task tool to delegate to critic agent with priority validation focus:\n\n**Priority Critic Prompt**: \"Analyze the priority classification using the BINARY CONFIDENCE SYSTEM validation:\n\n**VALIDATION CRITERIA**:\n- Verify 'Confident' claims only if ALL 6 strict criteria were genuinely met\n- Challenge confidence claims that lack sufficient evidence\n- Look for overlooked conflicting indicators (enhancement, nice to have, future, optional)\n- Validate precedent claims by checking cited similar issues\n- Assess whether reasoning is truly falsifiable with specific evidence\n- Question if 3+ keywords/indicators were actually present and relevant\n\n**VALIDATION ACTIONS**:\n- If 'Confident' claim is unjustified: Remove priority label (default to medium)\n- If 'Uncertain' but strong evidence exists: Consider adding priority label\n- Provide Priority Validation assessment\n- Adjust labels only when confidence assessment was incorrect\n\n**SKEPTICAL FOCUS**: Be particularly skeptical of 'Confident' claims - prevent priority inflation.\"\n\n### **Step 9: Create Final Issue**\nUse Task tool to delegate to github-issues-workflow agent:\n- Create GitHub issue with refined content and validated labels/priority\n- Add comprehensive analysis comments:\n - Content Quality Assessment with refinement summary\n - Label Assignment Rationale\n - Priority Analysis with confidence justification\n - Validation Results from critic reviews\n- Return issue number, URL, and quality assessment summary\n\n## Final Confirmation\n\nProvide comprehensive issue details with:\n- Direct GitHub issue link\n- Final content quality assessment\n- Label assignment summary \n- Priority classification with confidence level\n- Summary of all agent assessments and refinements", "variables": [], "notes": "Create new GitHub Issue with comprehensive quality validation through mandatory 8-step workflow. Tools: Task.", "metadata": { "source": { "provider": "github-claude-commands", "repository": "https://github.com/ondrasek/acforge-cli", "path": ".claude/commands/issue/create.md", "ref": "2af5c8e78205cc9aeadb82c437087f461aed3fa7", "url": "https://github.com/ondrasek/acforge-cli/blob/2af5c8e78205cc9aeadb82c437087f461aed3fa7/.claude/commands/issue/create.md", "key": "ondrasek/acforge-cli/.claude/commands/issue/create.md" } } }

create

About

GitHub Issue Creation - Mandatory 8-Step Quality Workflow

Mandatory 8-Step Workflow Instructions

Step 1: Draft Creation

Step 2: Refine Content

Step 3: Criticize Refinement

Step 4: Adjust Based on Criticism

Step 5: Label Assignment

Step 6: Prioritize with BINARY CONFIDENCE SYSTEM

Step 7: Criticize Labels

Step 8: Criticize Priority

Step 9: Create Final Issue

Final Confirmation

Install create and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Use it

Manifest

Reviews