RFC Creator

Overview

This public intake copy packages

packages/skills-catalog/skills/(creation)/create-rfc

https://github.com/tech-leads-club/agent-skills

Use it when the operator needs the upstream workflow, support files, and repository context to stay intact while the public validator and private enhancer continue their normal downstream flow.

This intake keeps the copied upstream files intact and uses

metadata.json

ORIGIN.md

RFC Creator You are an expert in creating Request for Comments (RFC) documents that clearly communicate proposals, capture alternatives considered, and drive structured decision-making across teams.

Imported source sections that did not map cleanly to the public headings are still preserved below or in the support files. Notable imported sections: Language Adaptation, RFC vs TDD, Document Structure, Section Templates, RFC Quality Checklist, Common Anti-Patterns to Avoid.

When to Use This Skill

Use this section as the trigger filter. It should make the activation boundary explicit before the operator loads files, runs commands, or opens a pull request.

• User asks to "write an RFC", "create an RFC", "draft a proposal", or "write a request for comments"
• User needs to propose a significant change and gather stakeholder feedback
• A major architectural, process, or product decision needs to be documented before acting
• User wants to align multiple teams or approvers before committing to a direction
• User asks to "document a decision" or "get buy-in" on a proposal
• User needs to compare options and record the chosen direction with rationale

Operating Table

metadata.json

ORIGIN.md

references/section-templates.md

README.md

## Related Skills

Workflow

This workflow is intentionally editorial and operational at the same time. It keeps the imported source useful to the operator while still satisfying the public intake standards that feed the downstream enhancer flow.

1. RFC title (clear, action-oriented)
2. Background / context (what is the current state and why this matters)
3. Driver (who is proposing / responsible for the decision)
4. Approver(s) (who needs to approve)
5. Impact level (HIGH / MEDIUM / LOW)
6. At least 1 explicit assumption (with confidence level)
7. At least 2 decision criteria (with weights), stated before options

Imported Workflow Notes

Imported: Interactive Workflow

Step 1: Gather Context (if not provided)

If the user provides no context, use AskQuestion to collect basic information:

{
  "title": "RFC Information",
  "questions": [
    {
      "id": "rfc_topic",
      "prompt": "What is the topic or change you want to propose?",
      "options": [
        { "id": "free_text", "label": "I'll describe it below" }
      ]
    },
    {
      "id": "rfc_impact",
      "prompt": "What is the estimated impact of this change?",
      "options": [
        { "id": "high", "label": "HIGH - affects multiple teams, systems, or users" },
        { "id": "medium", "label": "MEDIUM - affects one team or system" },
        { "id": "low", "label": "LOW - limited scope, easily reversible" }
      ]
    },
    {
      "id": "rfc_urgency",
      "prompt": "Is there a due date or urgency?",
      "options": [
        { "id": "urgent", "label": "Yes, we need a decision soon" },
        { "id": "planned", "label": "Part of planned roadmap" },
        { "id": "open", "label": "No fixed deadline" }
      ]
    },
    {
      "id": "rfc_options",
      "prompt": "Do you have options/alternatives in mind?",
      "options": [
        { "id": "yes", "label": "Yes, I have 2+ options to compare" },
        { "id": "one", "label": "I have a preferred option, need to document alternatives" },
        { "id": "no", "label": "No, need help structuring options" }
      ]
    }
  ]
}

Step 2: Validate Mandatory Fields

MANDATORY fields — ask if missing:

• RFC title (clear, action-oriented)
• Background / context (what is the current state and why this matters)
• Driver (who is proposing / responsible for the decision)
• Approver(s) (who needs to approve)
• Impact level (HIGH / MEDIUM / LOW)
• At least 1 explicit assumption (with confidence level)
• At least 2 decision criteria (with weights), stated before options
• At least 2 options considered (including "do nothing" when relevant)
• Recommended option with rationale tied back to the decision criteria

If any of these are missing, ask IN THE USER'S LANGUAGE before generating the document.

Step 3: Detect RFC Type and Tailor Sections

Step 4: Generate RFC Document

Generate the RFC in Markdown following the templates below.

Step 5: Offer Next Steps

After generating, offer:

RFC Created: "[Title]"

Sections included:
- Mandatory: Header & Metadata, Background, Assumptions, Decision Criteria, Options Considered, Action Items, Outcome
- Recommended: Relevant Data, Pros/Cons comparison, Cost estimate, Resources

Suggested next steps:
- Share with Contributors for feedback
- Set a decision deadline
- Schedule a review meeting with Approvers
- Link related Jira/Linear tickets

Would you like me to:
1. Add more options to compare?
2. Create a follow-up technical design doc (TDD) for implementation details?
3. Publish this to Confluence?

Imported: Language Adaptation

CRITICAL: Always generate the RFC in the same language as the user's request. Detect the language automatically and generate all content in that language.

• Keep technical terms in English when appropriate (e.g., "API", "RFC", "rollback", "stakeholder")
• Company/product names remain in original language
• Use natural, professional language for the target language

Examples

Example 1: Ask for the upstream workflow directly

Use @create-rfc to handle <task>. Start from the copied upstream workflow, load only the files that change the outcome, and keep provenance visible in the answer.

Explanation: This is the safest starting point when the operator needs the imported workflow, but not the entire repository.

Example 2: Ask for a provenance-grounded review

Review @create-rfc against metadata.json and ORIGIN.md, then explain which copied upstream files you would load first and why.

Explanation: Use this before review or troubleshooting when you need a precise, auditable explanation of origin and file selection.

Example 3: Narrow the copied support files before execution

Use @create-rfc for <task>. Load only the copied references, examples, or scripts that change the outcome, and name the files explicitly before proceeding.

Explanation: This keeps the skill aligned with progressive disclosure instead of loading the whole copied package by default.

Example 4: Build a reviewer packet

Review @create-rfc using the copied upstream files plus provenance, then summarize any gaps before merge.

Explanation: This is useful when the PR is waiting for human review and you want a repeatable audit packet.

Best Practices

Treat the generated public skill as a reviewable packaging layer around the upstream repository. The goal is to keep provenance explicit and load only the copied source material that materially improves execution.

• Keep the imported skill grounded in the upstream repository; do not invent steps that the source material cannot support.
• Prefer the smallest useful set of support files so the workflow stays auditable and fast to review.
• Keep provenance, source commit, and imported file paths visible in notes and PR descriptions.
• Point directly at the copied upstream files that justify the workflow instead of relying on generic review boilerplate.
• Treat generated examples as scaffolding; adapt them to the concrete task before execution.
• Route to a stronger native skill when architecture, debugging, design, or security concerns become dominant.

Troubleshooting

Problem: The operator skipped the imported context and answered too generically

Symptoms: The result ignores the upstream workflow in

packages/skills-catalog/skills/(creation)/create-rfc

metadata.json

ORIGIN.md

Problem: The imported workflow feels incomplete during review

Symptoms: Reviewers can see the generated

SKILL.md

Problem: The task drifted into a different specialization

Symptoms: The imported skill starts in the right place, but the work turns into debugging, architecture, design, security, or release orchestration that a native skill handles better. Solution: Use the related skills section to hand off deliberately. Keep the imported provenance visible so the next skill inherits the right context instead of starting blind.

Related Skills

• @accessibility

  - Use when the work is better handled by that native specialization after this imported skill establishes context.

• @ai-cold-outreach

  - Use when the work is better handled by that native specialization after this imported skill establishes context.

• @ai-pricing

  - Use when the work is better handled by that native specialization after this imported skill establishes context.

• @ai-sdr

  - Use when the work is better handled by that native specialization after this imported skill establishes context.

Additional Resources

Use this support matrix and the linked files below as the operator packet for this imported skill. They should reflect real copied source material, not generic scaffolding.

references

references/section-templates.md

examples

examples/n/a

scripts

scripts/n/a

agents

agents/n/a

assets

assets/n/a

• section-templates.md
• README.md
• section-templates.md

Imported Reference Notes

Imported: RFC vs TDD

Use RFC when the decision itself needs alignment. Use TDD when the decision is made and you need to document the implementation approach.

Imported: Document Structure

Mandatory Sections

1. Header & Metadata
2. Background
3. Assumptions
4. Decision Criteria
5. Options Considered (minimum 2)
6. Action Items
7. Outcome

Recommended Sections

8. Relevant Data — metrics, research, evidence
9. Pros and Cons (per option)
10. Estimated Cost (effort/complexity/monetary)
11. Resources — links, references, prior art

Imported: Section Templates

Read

references/section-templates.md

Imported: RFC Quality Checklist

Before finalizing, verify:

• Title: Clear, action-oriented, specific (not "RFC about the database")
• Impact: Assessed as HIGH / MEDIUM / LOW with justification
• Background: Current state + problem + why now + cost of inaction
• Assumptions: Explicit, with confidence levels and invalidation triggers
• Decision Criteria: Defined before options, with weights; Must-haves identified
• Data: At least some evidence supporting the need for change
• Options: Minimum 2 options (including "do nothing" for significant changes)
• Options evaluated against criteria: Not just pros/cons in isolation
• Pros/Cons: Honest assessment, not just selling one option
• Cost: Effort estimate for each option (even if rough)
• RACI: Driver, Approver(s), Contributors, Informed all identified
• Action Items: Concrete next steps after the decision
• Outcome: Left as placeholder to be filled when decision is made

Imported: Common Anti-Patterns to Avoid

Predetermined Conclusion Disguised as RFC

BAD:

We should use Kubernetes. Here are some reasons. Option 2 is to not use Kubernetes (obviously wrong).

GOOD:

Option 1: Adopt Kubernetes — [genuine pros and cons]
Option 2: Stick with Docker Compose — [genuine pros and cons]
Option 3: Move to managed container platform (ECS/Cloud Run) — [genuine pros and cons]

Vague Background

BAD:

Our current deployment process has some issues.

GOOD:

Our current deployment process requires 45 minutes of manual steps and has caused 3 production incidents in the past quarter due to human error. The team spends ~8 hours/week on deployment-related tasks.

Missing "Do Nothing" Option

Always include the status quo as an option for significant changes — it forces honest evaluation of whether action is truly needed.

No Decision Criteria (or criteria defined after options)

BAD: Presenting options first, then listing criteria — which looks like the criteria were chosen to justify a preferred option.

GOOD: Define criteria with weights before listing options. Then evaluate each option against them explicitly. The recommendation section should reference which criteria drove the decision.

Hidden or Unstated Assumptions

BAD:

We'll migrate to the new system over 6 months.

GOOD:

Assumption: The team has 2 engineers available for migration work in Q3.
Confidence: Medium. Invalidated if Q3 headcount changes.

Unstated assumptions become invisible time bombs. When the RFC outcome stops working six months later, no one can tell whether the decision was wrong or whether a hidden assumption was invalidated.

Imported: Output Summary Format

After generating the RFC:

RFC Created: "[Title]"

Impact: HIGH / MEDIUM / LOW
Status: NOT STARTED

Sections included:
- Header & Metadata (Driver, Approver, Due Date)
- Background (current state, problem, why now)
- N options compared with pros/cons and cost estimates
- Action Items (M tasks identified)
- Outcome (placeholder — to be filled after decision)

Suggested next steps:
- Share with Contributors listed for feedback
- Set the decision meeting for [Due Date]
- Update Status to IN PROGRESS

Would you like me to add anything else?

Imported: Important Notes

• RFC is for decisions, not implementation — once the RFC is decided, create a TDD for the implementation plan
• Honest options are critical — a one-sided RFC undermines trust and produces bad decisions
• "Do nothing" is always an option — helps assess whether change is truly worth it
• Outcome section is filled after the fact — leave as placeholder during drafting
• Language adaptation — always write in the user's language
• Respect user's context — if the user provides rich context, use it; don't ask for what's already given
• Be concise in options — focus on the decision factors, not implementation details
• RFCs age — date everything; decisions made without context become confusing later