fix(providers): remove ~anthropic, skip ~ prefixes in sync script (#69)

* fix(providers): remove ~anthropic, skip ~ prefixes in sync script OpenRouter uses ~ prefixes for internal auto-routing aliases (e.g. ~anthropic). These are not real providers — they already route through openrouter.toml. The generated ~anthropic.toml was confusing (looked like a stale backup) and redundant with the existing openrouter provider. - Delete providers/~anthropic.toml - Skip provider IDs starting with ~ in sync-pricing.py --create-missing * fix(providers): remove morph, aider, kwaipilot - morph: specialized code-editing/patching tool, not a general LLM provider - aider: CLI meta-tool wrapper (base_url empty), redundant with claude-code/codex-cli/gemini-cli/qwen-code - kwaipilot: Kwai internal coding assistant routed via OpenRouter, niche * fix(sync): add morph/aider/kwaipilot to SKIP_PROVIDERS to prevent re-creation * feat(sync): merge OpenRouter-only providers into openrouter.toml Instead of generating standalone .toml files that just wrap the OpenRouter endpoint, merge their models directly into openrouter.toml with the standard 'openrouter/{provider}/{model}' ID convention. - Add _build_model_fields() and _model_lines() helpers to deduplicate model rendering between standalone and merged paths - Add merge_into_openrouter() that appends new models idempotently - generate_provider_toml() now only runs for providers in PROVIDER_API - --create-missing routes OpenRouter-only providers to merge_into_openrouter * fix(providers): remove 14 OpenRouter-only standalone files These providers have no direct public API and all route through openrouter.ai/api/v1. Per the new sync-pricing.py policy, their models will be merged into openrouter.toml on the next CI run instead of living in separate files that just wrap the OpenRouter endpoint. Removed: allenai, deepcogito, essentialai, inclusionai, inflection, liquid, meituan, nex-agi, nousresearch, prime-intellect, relace, switchpoint, tngtech, writer * fix(providers): remove 7 niche providers with no driver support No dedicated LLM driver code exists for these providers — they rely purely on OpenAI-compatible passthrough with no special handling. Removing them reduces registry noise; users can still reach them via openrouter.toml if needed. Removed: microsoft, ibm-granite, xiaomi, upstage, inception, aion-labs, arcee-ai * fix(providers): remove ai21, chutes, venice All three use ApiFormat::OpenAI with no special handling — pure passthrough. No registry entry needed; users can reach them via openrouter.toml or by adding a custom provider. * docs(providers): rewrite README with full provider catalog and inclusion criteria - List all 46 providers grouped by category with descriptions - Document why each provider exists (direct API, unique endpoint, dedicated driver, local, CLI) - Add inclusion criteria section explaining when to create standalone files vs merging into openrouter.toml - Document sync script routing logic - Update model counts: 49→46 providers, 339→232 models * docs: add comprehensive READMEs for all registry sections + deepinfra provider - agents/README.md: 32 agents across 7 categories with capability field reference - channels/README.md: 44 channels across 5 categories with protocol reference table - hands/README.md: 18 hands across 5 categories with HAND.toml format guide - mcp/README.md: 33 MCP servers across 5 categories with transport/auth format - plugins/README.md: 12 plugins with hook protocol documentation - skills/README.md: 60 skills across 9 categories with SKILL.md format guide - providers/deepinfra.toml: add DeepInfra serverless inference (5 models)
2026-04-24 00:02:33 +09:00
1 parent dbfb32d9d4
commit d43077afa9
@@ -1,70 +1,148 @@
-# MCP Servers
+# MCP Servers Registry

-MCP (Model Context Protocol) server templates for LibreFang. Each entry connects LibreFang to an external service (GitHub, Slack, databases, etc.).
+MCP (Model Context Protocol) servers connect LibreFang agents to external services. Each entry in this directory is a single `.toml` file describing an MCP server: how to launch it, what credentials it requires, and which category of functionality it provides.

-## Structure
+When an agent has an MCP server attached, the server's tools appear automatically in the agent's tool list alongside built-in tools like `file_read` and `web_search`.

-```
-mcp/
-├── github.toml
-├── slack.toml
-├── postgresql.toml
-└── ...
-```
+## File Format

-## MCP Server TOML Format
+Each MCP server is a flat `.toml` file. The filename (without `.toml`) must match the `id` field.

 ```toml
-id = "service-id"                 # Must match filename (without .toml)
-name = "Service Name"
-description = "What this MCP server provides"
-category = "devtools"             # devtools | communication | storage | monitoring | data
-icon = "🐙"
-tags = ["relevant", "tags"]
+id = "github"                         # must match filename
+name = "GitHub"
+description = "Access GitHub repos, issues, PRs, and organizations"
+category = "devtools"                 # devtools | data | cloud | communication | productivity | ai
+icon = "lucide:github"
+tags = ["git", "vcs", "code"]

-[transport]                       # MCP server transport config
-type = "stdio"                    # "stdio" or "sse"
+[transport]
+type = "stdio"                        # stdio | sse
 command = "npx"
-args = ["-y", "@pkg/mcp-server"]
+args = ["-y", "@modelcontextprotocol/server-github@2025.4.8"]

-[[required_env]]                  # Required environment variables
-name = "SERVICE_API_KEY"
-label = "API Key"
-help = "How to obtain this key"
+[[required_env]]                      # repeat block for each required credential
+name = "GITHUB_PERSONAL_ACCESS_TOKEN"
+label = "GitHub Personal Access Token"
+help = "A fine-grained or classic PAT with repo and read:org scopes"
 is_secret = true
-get_url = "https://..."
+get_url = "https://github.com/settings/tokens"

-[oauth]                           # Optional: OAuth config
+[oauth]                               # optional: OAuth flow config
 provider = "github"
-scopes = ["repo"]
-auth_url = "https://..."
-token_url = "https://..."
+scopes = ["repo", "read:org"]
+auth_url = "https://github.com/login/oauth/authorize"
+token_url = "https://github.com/login/oauth/access_token"

 [health_check]
 interval_secs = 60
 unhealthy_threshold = 3

 setup_instructions = """
-Step-by-step setup guide for users.
+1. Go to https://github.com/settings/tokens and create a Personal Access Token.
+2. Paste the token into the GITHUB_PERSONAL_ACCESS_TOKEN field.
+3. Alternatively, use the OAuth flow to authorize LibreFang directly.
 """
+
+[i18n.zh]
+name = "GitHub"
+description = "通过官方 MCP 服务器访问 GitHub 仓库、Issue、Pull Request 与组织。"
 ```

-## Current MCP Servers (25)
+## Installing an MCP Server

-| MCP Server | Category | Service |
-|------------|----------|---------|
-| github | devtools | GitHub repos, issues, PRs |
-| slack | communication | Slack messaging |
-| notion | productivity | Notion pages and databases |
-| postgresql | storage | PostgreSQL database |
-| ... | | See each file for details |
+```bash
+# List all available MCP servers
+librefang catalog mcp
+
+# Install an MCP server and attach it to an agent
+librefang mcp install github
+librefang mcp attach github --agent coder
+
+# Or specify the agent when installing
+librefang mcp install github --agent coder
+
+# Set required credentials
+librefang config set-env GITHUB_PERSONAL_ACCESS_TOKEN ghp_xxx
+
+# List attached MCP servers for an agent
+librefang mcp list --agent coder
+
+# Remove an MCP server from an agent
+librefang mcp detach github --agent coder
+```
+
+## All MCP Servers (33 total)
+
+### Development Tools (devtools)
+
+| ID | Name | Transport | Credentials Required | Description |
+|----|------|-----------|---------------------|-------------|
+| bitbucket | Bitbucket | stdio (npx) | `BITBUCKET_*` | Access Bitbucket repositories, pull requests, and pipelines |
+| fetch | Fetch | stdio (npx) | none | Fetch any web URL and receive its content as clean Markdown |
+| filesystem | Filesystem | stdio (npx) | none | Read, write, search, and manage local files |
+| git | Git | stdio (npx) | none | Inspect local Git repos — commits, diffs, branches, blame |
+| github | GitHub | stdio (npx) | `GITHUB_PERSONAL_ACCESS_TOKEN` | Access GitHub repos, issues, PRs, and organizations |
+| gitlab | GitLab | stdio (npx) | `GITLAB_*` | Access GitLab projects, MRs, issues, and CI/CD pipelines |
+| jira | Jira | stdio (npx) | `JIRA_*` | Access Jira issues, projects, boards, and sprints |
+| linear | Linear | stdio (npx) | `LINEAR_API_KEY` | Manage Linear issues, projects, cycles, and teams |
+| puppeteer | Puppeteer | stdio (npx) | none | Control headless Chrome — navigate, screenshot, scrape |
+| sentry | Sentry | stdio (npx) | `SENTRY_AUTH_TOKEN` | Monitor Sentry error tracking, issues, and releases |
+
+### Data
+
+| ID | Name | Transport | Credentials Required | Description |
+|----|------|-----------|---------------------|-------------|
+| elasticsearch | Elasticsearch | stdio (npx) | `ELASTICSEARCH_*` | Search and manage Elasticsearch indices and documents |
+| google-maps | Google Maps | stdio (npx) | `GOOGLE_MAPS_API_KEY` | Geocoding, directions, distance matrix, and place search |
+| mongodb | MongoDB | stdio (npx) | `MONGODB_CONNECTION_STRING` | Query and manage MongoDB databases and collections |
+| postgresql | PostgreSQL | stdio (npx) | `POSTGRES_CONNECTION_STRING` | Query and manage PostgreSQL databases |
+| redis | Redis | stdio (npx) | `REDIS_URL` | Access and manage Redis key-value stores |
+| sqlite-mcp | SQLite | stdio (npx) | none | Query and manage local SQLite databases |
+
+### Cloud
+
+| ID | Name | Transport | Credentials Required | Description |
+|----|------|-----------|---------------------|-------------|
+| aws | AWS | stdio (npx) | `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY` | Manage S3, EC2, Lambda, and other AWS resources |
+| azure-mcp | Azure | stdio (npx) | `AZURE_*` | Manage Azure VMs, Storage, and App Services |
+| gcp-mcp | GCP | stdio (npx) | `GOOGLE_APPLICATION_CREDENTIALS` | Manage GCP Compute, Cloud Storage, and BigQuery |
+
+### Communication
+
+| ID | Name | Transport | Credentials Required | Description |
+|----|------|-----------|---------------------|-------------|
+| discord-mcp | Discord | stdio (npx) | `DISCORD_TOKEN` | Access Discord servers, channels, and messages |
+| slack | Slack | stdio (npx) | `SLACK_BOT_TOKEN` | Access Slack channels, messages, and users |
+| teams-mcp | Microsoft Teams | stdio (npx) | `TEAMS_*` | Access Teams channels, chats, and messages |
+
+### Productivity
+
+| ID | Name | Transport | Credentials Required | Description |
+|----|------|-----------|---------------------|-------------|
+| dropbox | Dropbox | stdio (npx) | `DROPBOX_ACCESS_TOKEN` | Access and manage Dropbox files and folders |
+| gmail | Gmail | stdio (npx) | `GMAIL_*` | Read, send, and manage Gmail messages and drafts |
+| google-calendar | Google Calendar | stdio (npx) | `GOOGLE_*` | Manage Google Calendar events and availability |
+| google-drive | Google Drive | stdio (npx) | `GOOGLE_*` | Browse, search, and read files from Google Drive |
+| notion | Notion | stdio (npx) | `NOTION_API_KEY` | Access and manage Notion pages, databases, and blocks |
+| time | Time | stdio (npx) | none | Get current time, convert timezones, calculate differences |
+| todoist | Todoist | stdio (npx) | `TODOIST_API_TOKEN` | Manage Todoist tasks, projects, and labels |
+
+### AI
+
+| ID | Name | Transport | Credentials Required | Description |
+|----|------|-----------|---------------------|-------------|
+| brave-search | Brave Search | stdio (npx) | `BRAVE_API_KEY` | Perform web searches using the Brave Search API |
+| exa-search | Exa Search | stdio (npx) | `EXA_API_KEY` | AI-powered neural search and web content retrieval |
+| memory | Memory | stdio (npx) | none | Persistent knowledge graph for facts and relationships across sessions |
+| sequential-thinking | Sequential Thinking | stdio (npx) | none | Structured multi-step reasoning with revisable thought chains |

 ## Adding a New MCP Server

-1. Create `mcp/<name>.toml`
-2. Ensure `id` matches the filename
-3. Test the MCP server command locally
-4. Run `python scripts/validate.py`
-5. Submit a PR
+1. Create `mcp/<name>.toml` — the filename must match the `id` field.
+2. Test the transport command locally: run `npx -y <package>` and verify it starts without errors.
+3. List all `[[required_env]]` entries so the UI can prompt users for credentials.
+4. Run `python scripts/validate.py`.
+5. Submit a PR.

 See [CONTRIBUTING.md](../CONTRIBUTING.md) for the full guide.