chore(workflows): move agent workflows to global dir

This commit is contained in:
diegosouzapw
2026-02-21 09:11:14 -03:00
parent 340dcf9515
commit 1aa27ceefe
3 changed files with 0 additions and 160 deletions

View File

@@ -1,54 +0,0 @@
---
description: Git workflow — NEVER commit directly to main. Always use feature branches.
---
# Git Workflow
## ⚠️ CRITICAL RULE: NEVER commit directly to `main`
## Steps
1. **Before starting any work**, create a feature branch from `main`:
```bash
git checkout main && git pull origin main
git checkout -b feature/<feature-name>
```
2. **During development**, commit to the feature branch:
```bash
git add -A && git commit -m "<type>(<scope>): <description>"
```
3. **Before pushing**, verify the build passes:
```bash
npm run build
```
4. **When the feature is complete and verified**, push the branch and STOP:
```bash
git push origin feature/<feature-name>
```
5. **DO NOT** create a PR, merge, or push to `main`. Let the user handle that.
## Branch naming convention
- `feature/<name>` — new features
- `fix/<name>` — bugfixes
- `refactor/<name>` — refactoring
- `docker/<name>` — Docker / infrastructure changes
- `style/<name>` — UI / CSS changes
## Commit types
- `feat` — new feature
- `fix` — bugfix
- `refactor` — code refactoring
- `style` — UI / CSS changes
- `docker` — Docker / infrastructure
- `docs` — documentation
- `chore` — maintenance

View File

@@ -1,65 +0,0 @@
---
description: How to respond to GitHub issues with insufficient information
---
# GitHub Issue Triage — Requesting More Information
When analyzing a GitHub issue and the provided information is insufficient to diagnose or implement a fix, follow these steps:
## 1. Identify Missing Information
Check if the issue contains:
- [ ] Clear description of the problem or feature
- [ ] Steps to reproduce (for bugs)
- [ ] Error messages / logs
- [ ] Environment details (OS, version, deployment method)
- [ ] Expected vs actual behavior
- [ ] Screenshots (if UI-related)
## 2. Draft a Response (in English)
Use this template structure, adapting to what's specifically missing:
```markdown
Thanks for reporting this issue! 🙏
To help us investigate and fix this, could you please provide some more details?
1. **Environment**: What OS are you running? How did you install OmniRoute (npm global, Docker, from source)?
2. **Steps to Reproduce**: What exact steps lead to this issue?
3. **Error Logs**: Can you share the full server logs? (Check the terminal or `logs/application/app.log`)
4. **Expected vs Actual**: What did you expect to happen, and what actually happened?
5. **Screenshots**: If this is a UI issue, a screenshot would be very helpful.
This information will help us reproduce and fix the issue more quickly. Thank you!
```
## 3. Adapt the Template
- Remove items that are already provided in the issue
- Add domain-specific questions based on the issue type:
- **OAuth/Auth issues**: Ask for provider name, auth flow used, token errors
- **Multi-account issues**: Ask how many accounts, which providers, dashboard screenshots
- **Startup/crash issues**: Ask for full startup logs, OS version, Node.js version
- **UI rendering issues**: Ask for browser name, screen resolution, screenshots
- **API issues**: Ask for request/response examples, curl commands used
## 4. Post the Response
// turbo
Post the response as a comment on the GitHub issue using the GitHub CLI:
```bash
gh issue comment <ISSUE_NUMBER> --repo diegosouzapw/OmniRoute --body "<RESPONSE_BODY>"
```
If `gh` CLI is not available, provide the response text to the user to post manually.
## 5. Add Labels (optional)
If applicable, add a `needs-info` label:
```bash
gh issue edit <ISSUE_NUMBER> --repo diegosouzapw/OmniRoute --add-label "needs-info"
```

View File

@@ -1,41 +0,0 @@
---
description: How to automatically summarize recent changes and update README and CHANGELOG
---
# Update Documentation and Changelog Workflow
When asked to run the `/update-docs` workflow or summarize recent changes into the documentation, follow these steps:
## 1. Gather Recent Work Context
1. **Review Task History**: Check the current `task.md`, `walkthrough.md`, and any `implementation_plan.md` in the agent's brain directory to understand exactly what features, bug fixes, and architectural changes were recently made.
2. **Identify Key Features**: Group the changes into logical categories:
- **New Features** (e.g. new settings, new endpoints, new UI components)
- **Bug Fixes** (e.g. layout fixes, crash resolutions, cross-platform fixes)
- **Security/Performance** (e.g. auth gating, new optimizations)
## 2. Update the CHANGELOG.md
1. If `CHANGELOG.md` does not exist in the root directory, create it.
2. Under the `## [Unreleased]` section (or create a new version header if releasing):
- Add a `### Added` section for new features.
- Add a `### Fixed` section for bug fixes.
- Summarize the recent work into bullet points, referencing GitHub issue numbers if applicable (e.g. `Fix Windows server startup crash (Issue #98)`).
## 3. Update README.md (and translated versions like README_pt-BR.md)
1. **Main Features List**: If a notable new feature was added (like a new security setting or API option), add it to the feature list section.
2. **Configuration/Settings**: If any new environment variables or dashboard settings were introduced, document them in the configuration or usage sections so users know how to enable them.
3. If there are translated versions of the README (like Portuguese), ensure those are updated equivalently.
## 4. Update Technical Docs (docs/ directory)
1. Use `list_dir` to inspect the `docs/` folder.
2. Determine which technical documents need updates. For example:
- If an API changed, update `docs/API_REFERENCE.md` or similar.
- If security settings changed, update `docs/SECURITY.md` or `docs/CONFIGURATION.md`.
3. Read the relevant files, update their content using `replace_file_content`, and briefly explain how the new feature works under the hood.
## 5. Verify the Updates
Ensure all markdown files are properly formatted and no existing sections were accidentally deleted. Present the changes to the user showing which files were updated.