What is a SKILL.md file?

A SKILL.md is a small markdown file that tells an AI agent how to perform a specific task well. It has a name, a description, and a body the agent reads as expert guidance. Anthropic introduced the format for Claude Agent Skills; the broader open-source community now publishes thousands of them on GitHub for domains like API integration, frontend design, security review, video rendering, and more.

How does my AI agent use Skillhound?

Three ways. The MCP server gives Claude Code, Cursor, Claude Desktop, Codex, Cline, Windsurf, and Manus a search_skills tool that fans out across the index proactively. The CLI lets any shell-based agent search from the terminal. The REST API gives custom integrations the same data with a Bearer token. The web UI is free; programmatic surfaces are 25 free agent searches per month, then $20/month unlimited.

How often is the index refreshed?

Every 48 hours. A scheduled ECS Fargate task discovers new SKILL.md files via the GitHub Code Search API, expands repository trees, deduplicates by content hash, and reloads OpenSearch. The full corpus is currently around 135K skills.

How are skills ranked?

Default sort is by GitHub stars on the parent repository, descending. Highest-starred skills from recognized organizations bubble up first. You can also sort by recently-updated to surface fresh skills before they've gathered stars.

Is Skillhound compatible with ChatGPT?

ChatGPT does not yet support MCP servers, so the MCP install path doesn't apply. You can still use the web UI, copy any skill's markdown, and paste it into a ChatGPT system prompt or use the 'Open in ChatGPT' button on any skill page. Claude, Cursor, Claude Desktop, Codex, Cline, Windsurf, and Manus all support MCP and can hook into Skillhound directly.

What is a SKILL.md file?

A SKILL.md is a small markdown file that tells an AI agent how to perform a specific task well. It has a name, a description, and a body the agent reads as expert guidance. Anthropic introduced the format for Claude Agent Skills; the broader open-source community now publishes thousands of them on GitHub for domains like API integration, frontend design, security review, video rendering, and more.

How does my AI agent use Skillhound?

Three ways. The MCP server gives Claude Code, Cursor, Claude Desktop, Codex, Cline, Windsurf, and Manus a search_skills tool that fans out across the index proactively. The CLI lets any shell-based agent search from the terminal. The REST API gives custom integrations the same data with a Bearer token. The web UI is free; programmatic surfaces are 25 free agent searches per month, then $20/month unlimited.

How often is the index refreshed?

Every 48 hours. A scheduled ECS Fargate task discovers new SKILL.md files via the GitHub Code Search API, expands repository trees, deduplicates by content hash, and reloads OpenSearch. The full corpus is currently around 135K skills.

How are skills ranked?

Default sort is by GitHub stars on the parent repository, descending. Highest-starred skills from recognized organizations bubble up first. You can also sort by recently-updated to surface fresh skills before they've gathered stars.

Is Skillhound compatible with ChatGPT?

ChatGPT does not yet support MCP servers, so the MCP install path doesn't apply. You can still use the web UI, copy any skill's markdown, and paste it into a ChatGPT system prompt or use the 'Open in ChatGPT' button on any skill page. Claude, Cursor, Claude Desktop, Codex, Cline, Windsurf, and Manus all support MCP and can hook into Skillhound directly.

What is a SKILL.md file?

A SKILL.md is a small markdown file that tells an AI agent how to perform a specific task well. It has a name, a description, and a body the agent reads as expert guidance. Anthropic introduced the format for Claude Agent Skills; the broader open-source community now publishes thousands of them on GitHub for domains like API integration, frontend design, security review, video rendering, and more.

How does my AI agent use Skillhound?

Three ways. The MCP server gives Claude Code, Cursor, Claude Desktop, Codex, Cline, Windsurf, and Manus a search_skills tool that fans out across the index proactively. The CLI lets any shell-based agent search from the terminal. The REST API gives custom integrations the same data with a Bearer token. The web UI is free; programmatic surfaces are 25 free agent searches per month, then $20/month unlimited.

How often is the index refreshed?

Every 48 hours. A scheduled ECS Fargate task discovers new SKILL.md files via the GitHub Code Search API, expands repository trees, deduplicates by content hash, and reloads OpenSearch. The full corpus is currently around 135K skills.

How are skills ranked?

Default sort is by GitHub stars on the parent repository, descending. Highest-starred skills from recognized organizations bubble up first. You can also sort by recently-updated to surface fresh skills before they've gathered stars.

Is Skillhound compatible with ChatGPT?

ChatGPT does not yet support MCP servers, so the MCP install path doesn't apply. You can still use the web UI, copy any skill's markdown, and paste it into a ChatGPT system prompt or use the 'Open in ChatGPT' button on any skill page. Claude, Cursor, Claude Desktop, Codex, Cline, Windsurf, and Manus all support MCP and can hook into Skillhound directly.

goca-helper-skill · sazardev

\n- PascalCase for fields (e.g., `Email`, `CreatedAt`)\n- Length: 1-50 characters per field name\n\n**Auto-generated fields:**\n- `ID` (uint, primary key, auto-increment) — always added\n- `CreatedAt`, `UpdatedAt` (time.Time) — with `--timestamps`\n- `DeletedAt` (gorm.DeletedAt) — with `--soft-delete`\n\n---\n\n## Error Tracing \u0026 Debugging\n\n### Common errors and solutions\n\n| Error | Cause | Fix |\n|-------|-------|-----|\n| `--fields flag is required` | Missing required fields | Add `--fields \"Name:type\"` |\n| `invalid entity name` | Name doesn't match regex | Use PascalCase alphanumeric only (`^[A-Za-z][A-Za-z0-9]* ) |\n| `invalid field format: ...` | Wrong field syntax | Use `Name:type` pairs separated by commas |\n| `file already exists` | Conflict with existing file | Add `--force` to overwrite, or `--backup` to backup first |\n| `entity already exists` | Duplicate entity name | Use different name or `--force` |\n| `unknown database type` | Invalid `--database` | Use one of: postgres, postgres-json, mysql, mongodb, sqlite, sqlserver, elasticsearch, dynamodb |\n| `unknown handler type` | Invalid `--type` | Use one of: http, grpc, cli, worker, soap |\n| Go build fails after generation | Missing deps or imports | Run `go mod tidy` then `go build ./...` |\n\n### Debug workflow\n```bash\ngoca doctor # 1. Check project health\ngoca doctor --fix # 2. Auto-fix missing dirs\ngoca analyze # 3. Deep analysis\ngoca analyze --quality # or just quality checks\ngoca analyze --security # or just security checks\ngoca analyze --output json # machine-readable output\n```\n\n### Safety first (always available on generation commands)\n```bash\ngoca feature User --fields \"...\" --dry-run # preview without creating\ngoca feature User --fields \"...\" --force # overwrite existing\ngoca feature User --fields \"...\" --backup # backup before overwrite\ngoca feature User --fields \"...\" --dry-run --force --backup # combine safely\n```\n\n---\n\n## Flag Reference by Command\n\n### `goca feature \u003cname\u003e` — Full feature generation\n| Flag | Shorthand | Default | Description |\n|------|-----------|---------|-------------|\n| `--fields` | `-f` | **required** | Entity fields |\n| `--database` | `-d` | `postgres` | postgres, mysql, mongodb, sqlite, sqlserver, elasticsearch, dynamodb |\n| `--handlers` | | `http` | http, grpc, cli, worker, soap |\n| `--validation` | | `false` | Include validations |\n| `--business-rules` | `-b` | `false` | Business rule methods |\n| `--integration-tests` | | `false` | Generate integration tests |\n| `--mocks` | | `false` | Generate testify mocks |\n| `--cache` | `-c` | `false` | Redis cache decorator |\n| `--dry-run` | | `false` | Preview only |\n| `--force` | | `false` | Overwrite existing |\n| `--backup` | | `false` | Backup before write |\n\n### `goca entity \u003cname\u003e` — Domain entity\n| Flag | Shorthand | Default | Description |\n|------|-----------|---------|-------------|\n| `--fields` | `-f` | **required** | Entity fields |\n| `--validation` | | `false` | Include Validate() method |\n| `--business-rules` | `-b` | `false` | Business rules (IsAdult, IsExpensive, etc.) |\n| `--timestamps` | `-t` | `false` | CreatedAt + UpdatedAt |\n| `--soft-delete` | `-s` | `false` | DeletedAt field |\n| `--tests` | | `true` | Unit tests |\n| `--dry-run` / `--force` / `--backup` | | `false` | Safety flags |\n\n### `goca usecase \u003cname\u003e` — Business logic\n| Flag | Shorthand | Default | Description |\n|------|-----------|---------|-------------|\n| `--entity` | `-e` | **required** | Associated entity name |\n| `--operations` | `-o` | `create,read` | create, read, update, delete, list |\n| `--dto-validation` | `-d` | `false` | DTOs with validation |\n| `--async` | `-a` | `false` | Async operations |\n\n### `goca handler \u003centity\u003e` — Protocol adapter\n| Flag | Shorthand | Default | Description |\n|------|-----------|---------|-------------|\n| `--type` | `-t` | `http` | http, grpc, cli, worker, soap |\n| `--validation` | | `false` | Input validation |\n| `--swagger` | `-s` | `false` | Swagger docs (HTTP only) |\n\n### `goca repository \u003centity\u003e` — Persistence\n| Flag | Shorthand | Default | Description |\n|------|-----------|---------|-------------|\n| `--database` | `-d` | `\"\"` | postgres, mysql, mongodb, sqlite, sqlserver, elasticsearch, dynamodb |\n| `--interface-only` | `-i` | `false` | Interface only |\n| `--cache` | `-c` | `false` | Cache layer |\n| `--transactions` | `-t` | `false` | Transaction support |\n\n### `goca di` — Dependency injection\n| Flag | Shorthand | Default | Description |\n|------|-----------|---------|-------------|\n| `--features` | `-f` | **required** | Comma-separated feature names |\n| `--database` | `-d` | `postgres` | Database type |\n| `--wire` | `-w` | `false` | Use Google Wire |\n\n### `goca middleware \u003cname\u003e` — HTTP middleware\n| Flag | Default | Description |\n|------|---------|-------------|\n| `--types` | `cors,logging,recovery` | cors, logging, auth, rate-limit, recovery, request-id, timeout |\n\n---\n\n## Template System\n\n### Built-in templates (in `cmd/templates.go`):\n- `entityTemplate` — Entity struct + validation + methods\n- `useCaseTemplate` — Interface + service + DTOs\n- `repositoryTemplate` — Interface + DB implementation\n- `handlerTemplate` — HTTP handler with CRUD\n\n### Custom templates:\n```bash\ngoca template init # creates .goca/templates/ with defaults\ngoca template list # shows available custom templates\n```\nCustom `.tmpl` / `.tpl` files override built-in templates.\n\n### Template functions available:\n`title`, `lower`, `upper`, `camel` / `toCamelCase`, `pascal` / `toPascalCase`, `snake` / `toSnakeCase`, `kebab` / `toKebabCase`, `plural` / `toPlural`, `singular` / `toSingular`, `join`, `split`, `contains`, `hasPrefix`, `hasSuffix`, `trimSpace`, `replace`, `replaceAll`\n\n---\n\n## MCP Integration (AI Tool Calling)\n\nGoca can run as an MCP server for AI assistants:\n```bash\ngoca mcp-server --print-config vscode # VS Code config snippet\ngoca mcp-server --print-config claude # Claude Desktop config\ngoca mcp-server --print-config cursor # Cursor config\ngoca mcp-server --print-config zed # Zed config\n```\n\n### Available MCP tools (5 core + 11 utility):\n\n**Core generation tools (auto-adds `--no-interactive`):**\n- `goca_feature` — name, fields, database, validation, business_rules, handlers, integration_tests, mocks, dry_run, force\n- `goca_entity` — name, fields, validation, business_rules, timestamps, soft_delete, tests, dry_run, force\n- `goca_usecase` — name, fields, validation, dry_run, force\n- `goca_repository` — name, database, dry_run, force\n- `goca_handler` — name, type, validation, dry_run, force\n\n**Utility tools:**\n`goca_di`, `goca_integrate`, `goca_interfaces`, `goca_messages`, `goca_mocks`, `goca_init`, `goca_doctor`, `goca_upgrade`, `goca_ci`, `goca_middleware`, `goca_analyze`\n\n---\n\n## Quick Reference Card\n\n### Most common command patterns\n```bash\n# New project\ngoca init \u003cname\u003e -m \u003cmodule\u003e\n\n# Quick full feature\ngoca feature \u003cName\u003e -f \"field:type,field2:type2\" -d \u003cdb\u003e\n\n# Add to existing\ngoca entity \u003cName\u003e -f \"...\"\ngoca integrate --all\n\n# Validate\ngoca doctor \u0026\u0026 goca analyze\n\n# Safety\n\u003ccommand\u003e --dry-run # preview\n\u003ccommand\u003e --dry-run --force --backup # safe overwrite preview\n```\n\n### Naming conventions\n- Entities: PascalCase (`User`, `ProductOrder`)\n- Files: snake_case (`user.go`, `product_order.go`)\n- Packages: lowercase (`domain`, `usecase`)\n\n### Database types available\n`postgres`, `postgres-json`, `mysql`, `mongodb`, `sqlite`, `sqlserver`, `elasticsearch`, `dynamodb`\n\n### Handler types available\n`http`, `grpc`, `cli`, `worker`, `soap`\n\n### Operations for usecase\n`create`, `read`, `update`, `delete`, `list`\n","repo_fullName":"sazardev/goca","repo_stars":200,"repo_language":"Go","repo_license":"MIT","repo_pushedAt":"2026-05-26T19:46:35Z","owner_login":"sazardev","owner_type":"User","owner_name":"Omar Flores Salazar","owner_avatarUrl":"https://avatars.githubusercontent.com/u/66286300?v=4"}};