mcp-openapi-proxy
mcp-openapi-proxy is a Python package that implements a Model Context Protocol (MCP) server, designed to dynamically expose REST APIs—defined by OpenAPI specifications—as MCP tools. This facilitates seamless integration of OpenAPI-described APIs into MCP-based workflows.
v0.4.0 — tagged September 6, 2026; GitHub Release published October 15, 2026 (AEDT): MCP SDK 2.x / protocol 2026-07-28, with native stateless Streamable HTTP and legacy stdio support.
What's New in 0.4.0
MCP 2026-07-28 (SDK 2.x). The proxy is dual-stack: modern clients send
self-contained requests (no initialize, no Mcp-Session-Id) and list results
carry ttlMs / cacheScope. Legacy stdio clients that still handshake keep
working. Native Streamable HTTP is available with MCP_TRANSPORT=streamable-http
and is stateless — round-robin load balancing needs no sticky sessions.
See docs/MIGRATION-0.4.md for breaking changes, new
env vars, and gateway cutover notes. 0.3.4 remains the last release on
mcp 1.x (mcp>=1.2.0,<2).
What's New in 0.2.0
Works with every modern MCP-enabled client we tested. Strict MCP clients can now discover and call tools — the low-level server advertises correct capabilities and no longer crashes during resource/prompt discovery, and a slow spec download no longer crash-loops short-timeout clients. Verified live against the full list of mainstream agent CLIs:
- ✅ Codex, Gemini, Qwen, Kilocode, opencode — native tool calls over stdio
- ✅ Vibe — native discovery and read calls (writes were CLI-flaky, not a proxy issue)
- ✅ Letta — Cloud (via a remote streamable-HTTP MCP URL) and self-hosted (via stdio)
See the client matrix for attach mechanisms, models, and exact results.
> 📄 Full write-up: Verification case study — what the proxy is, the API + client matrices, and every defect found & fixed.
Prompts and resources are real now — including custom resources. Both MCP surfaces are functional and tested: the summarize_spec / whimsical_blog prompts and the spec_file resource, plus a new ADDITIONAL_RESOURCES env var that serves your own use-case documents (e.g. a NetBox naming policy or an Asana project-layout guide) as MCP resources — see examples/resources/.
Bug fixes (every one live-verified):
- MCP client discovery: empty capability set + a crash in resource discovery left strict clients seeing zero tools (#23) — fixed, with a full stdio-handshake test harness.
- IGNORE_SSL_TOOLS was ignored by the low-level dispatcher (#14) — fixed (original patch by @robbycochran, #15).
- Server crash-loop when a slow spec fetch outran a client's connect timeout (#28) — handshake now answers immediately, spec loads lazily, closed streams exit cleanly.
- API_AUTH_TYPE custom schemes (e.g. NetBox Token) sent no auth header at all (#24) — fixed.
- TOOL_WHITELIST never matched Slack-style dot paths like /users.list (#27) — fixed.
- TOOL_NAME_MAX_LENGTH was not respected, and name-truncation collisions silently dropped tools (#11) — fixed.
- Array parameters were emitted without items, which the OpenAI API rejects (#16) — fixed.
- EXTRA_HEADERS now accepts a JSON array and literal \n separators, not just real newlines (#17).
- Dead Render spec URL (#26) and incomplete ElevenLabs example (#29) — fixed; the GetZep example is documented for self-hosted Zep CE since the hosted endpoint now 401s (#38).
- Added a Dockerfile + glama.json for the Glama listing (#13); collapsible README examples + a verified-client/API matrix (#35).
Full environment-variable reference is in Environment Variables.
Table of Contents
- Overview
- Features
- Installation
- MCP Ecosystem Integration
- Modes of Operation
- FastMCP Mode (Simple Mode)
- Low-Level Mode (Default)
- Environment Variables
- Verified Clients & Live Results (2026-06-12)
- Tips
- Examples — Glama, Fly.io, Render, Slack, GetZep, Virustotal, Notion, Asana, APIs.guru, NetBox, Box, WolframAlpha, WordPress (collapsed)
- Troubleshooting
- License
Overview
The package offers two operational modes:
- Low-Level Mode (Default): Dynamically registers tools corresponding to all valid API endpoints specified in an OpenAPI document (e.g.
/chat/completionsbecomeschat_completions()). - FastMCP Mode (Simple Mode): Provides a streamlined approach by exposing a predefined set of tools (e.g.
list_functions()andcall_function()) based on static configurations.
Features
- Dynamic Tool Generation: Automatically creates MCP tools from OpenAPI endpoint definitions.
- Tool Annotations: Every OpenAPI-derived tool includes MCP
annotations(title, plusreadOnlyHint/idempotentHintfor GET and clearly read-ish POST, ordestructiveHintfor other methods) so clients such as Gemini Spark see the same fingerprint as Notion/Linear. - Simple Mode Option: Offers a static configuration alternative via FastMCP mode.
- OpenAPI Specification Support: Compatible with OpenAPI v3 with potential support for v2.
- Flexible Filtering: Allows endpoint filtering through whitelisting by paths or other criteria.
- Payload Authentication: Supports custom authentication via JMESPath expressions (e.g. for APIs like Slack that expect tokens in the payload not the HTTP header).
- Header Authentication: Uses
Bearerby default forAPI_KEYin the Authorization header, customizable for APIs like Fly.io requiringApi-Key. - MCP Integration: Seamlessly integrates with MCP ecosystems for invoking REST APIs as tools.
Installation
Install the package directly from PyPI using the following command:```bash
uvx mcp-openapi-proxy
### MCP Ecosystem Integration
To incorporate **mcp-openapi-proxy** into your MCP ecosystem configure it within your `mcpServers` settings. Below is a generic example:```json
{
"mcpServers": {
"mcp-openapi-proxy": {
"command": "uvx",
"args": ["mcp-openapi-proxy"],
"env": {
"OPENAPI_SPEC_URL": "${OPENAPI_SPEC_URL}",
"API_KEY": "${API_OPENAPI_KEY}"
}
}
}
}
Refer to the Examples section below for practical configurations tailored to specific APIs.
Modes of Operation
FastMCP Mode (Simple Mode)
- Enabled by: Setting the environment variable
OPENAPI_SIMPLE_MODE=true. - Description: Exposes a fixed set of tools derived from specific OpenAPI endpoints as defined in the code.
- Configuration: Relies on environment variables to specify tool behavior.
Low-Level Mode (Default)
- Description: Automatically registers all valid API endpoints from the provided OpenAPI specification as individual tools.
- Tool Naming: Derives tool names from normalized OpenAPI paths and methods.
- Behavior: Generates tool descriptions from OpenAPI operation summaries and descriptions.
Environment Variables
OPENAPI_SPEC_URL: (Required) The URL to the OpenAPI specification JSON file (e.g.https://example.com/spec.jsonorfile:///path/to/local/spec.json).OPENAPI_SIMPLE_MODE: (Optional) Set totrueto enable FastMCP mode.TOOL_WHITELIST: (Optional) A comma-separated list of endpoint paths to expose as tools.ADDITIONAL_RESOURCES: (Optional) Comma-separatedname=/path/to/fileentries served as extra MCP resources (use-case docs such as naming policies or layout conventions; seeexamples/resources/).TOOL_NAME_PREFIX: (Optional) A prefix to prepend to all tool names.API_KEY: (Optional) Authentication token for the API sent asBearerin the Authorization header by default.API_AUTH_TYPE: (Optional) Overrides the defaultBearerAuthorization scheme.api-keysends the key in the header named byAPI_AUTH_HEADER; any other value is used as a custom scheme prefix (e.g.Tokenfor NetBox →Authorization: Token).STRIP_PARAM: (Optional) JMESPath expression to strip unwanted parameters (e.g.tokenfor Slack).DEBUG: (Optional) Enables verbose debug logging when set to "true", "1", or "yes". Native Streamable HTTP also emits one INFO line per inbound JSON-RPC method (mcp rpc method=tools/list,mcp rpc method=tools/call name=…) without arguments or Authorization.EXTRA_HEADERS: (Optional) One or more outgoing HTTP headers. Accepts a JSON array (["X-A: 1","X-B: 2"]), oneHeader: Valueper line, or literal\n-separated entries (for configs that cannot embed newlines).SERVER_URL_OVERRIDE: (Optional) Overrides the base URL from the OpenAPI specification when set, useful for custom deployments.TOOL_NAME_MAX_LENGTH: (Optional) Truncates tool names to a max length.- Additional Variable:
OPENAPI_SPEC_URL_– a variant for unique per-test configurations (falls back toOPENAPI_SPEC_URL). IGNORE_SSL_SPEC: (Optional) Set totrueto disable SSL certificate verification when fetching the OpenAPI spec.IGNORE_SSL_TOOLS: (Optional) Set totrueto disable SSL certificate verification for API requests made by tools.API_AUTH_HEADER: (Optional) Header name used whenAPI_AUTH_TYPE=api-key(e.g.x-apikeyfor VirusTotal,xi-api-keyfor ElevenLabs). Defaults toAuthorization.OPENAPI_SPEC_FORMAT: (Optional) Set toyamlto parsefile://specs as YAML (remote specs auto-detect). Defaultjson.OPENAPI_SPEC_CACHE_TTL_SECONDS: (Optional) Live-first disk cache for remote specs: the cached copy is served only when the live fetch fails or stalls (and respawned servers fail fast to it). Default86400; set0to disable.ENABLE_TOOLS/ENABLE_PROMPTS/ENABLE_RESOURCES: (Optional) Feature gates for the three MCP surfaces in low-level mode; each defaults to enabled. Disabling a feature removes its handlers and its capability advertisement.PROMPT: (Optional) JSON object that registers one extra MCP prompt next tosummarize_specandwhimsical_blog, in both low-level and simple mode. Example:{"name":"asana_prompt","description":"Asana helper","messages":[{"user":"What is Asana?","assistant":"Asana is a work management platform."}]}. Each message may includeuser,assistant, or both. Invalid JSON is logged and ignored.summarize_specandwhimsical_blogare reserved names. Restart the server after changing it. Low-level mode also requires prompts to stay enabled (ENABLE_PROMPTS, default on). Custom prompt content is returned to connected MCP clients: do not put credentials or other private data inPROMPT, and protect public HTTP ingress with authentication and host/origin controls.CAPABILITIES_TOOLS/CAPABILITIES_PROMPTS/CAPABILITIES_RESOURCES: (Optional) AdvertiselistChangedon the corresponding capability (for clients that key on it). Defaultfalse.MCP_SERVER_NAME: (Optional) Identity advertised inserverInfo/server/discover. Set this to the proxied API (e.g.gpt-terminal-plus,flyio) so multiple instances are not all calledopenapi-proxy. Falls back to the OpenAPI file stem or spec-URL host, thenopenapi-proxy.MCP_SERVER_TITLE/MCP_SERVER_DESCRIPTION: (Optional) Display title and description alongside the name.MCP_TRANSPORT: (Optional)stdio(default) orstreamable-httpfor native stateless Streamable HTTP (MCP 2026-07-28; noMcp-Session-Id).MCP_HOST/MCP_PORT/MCP_PATH: (Optional) HTTP bind address (127.0.0.1), port (8000), and path (/mcp) whenMCP_TRANSPORT=streamable-http. Native Streamable HTTP also servesGET /healthz→{"ok":true,"name":,"port":}without taking the MCP request lock (no spec fetch, noinitialize).MCP_JSON_RESPONSE: (Optional)true(default) returns JSON instead of SSE on Streamable HTTP.MCP_LIST_TTL_MS/MCP_READ_TTL_MS: (Optional)ttlMsfor list results (default60000) andresources/read(default5000).MCP_ALLOWED_HOSTS: (Optional) Comma-separated Host allow-list for DNS-rebinding protection. Default*(protection off — typical behind nginx).-
MCP_REQUEST_STATE_KEY: (Optional) Shared HMAC key so MRTRrequestStateverifies across instances. Unset = process-local.
Unreleased onmain(not in tagged/PyPI v0.4.0): The next two raw-command logging settings are available only in newer source. -
MCP_LOG_RAW_COMMANDS: (Optional, default off)1/true/yesadds one extra INFO record per shell-execution tool call, emitted before the command runs:text mcp raw-command tool=post_command_execute_shell cd /srv/app && npm run build # <- the verbatim command, unmodified
The command is logged byte-for-byte — whitespace, blank lines, tabs and quoting are preserved; nothing is normalised, reconstructed, or redacted. The existing mcp rpc method=… name=… line is unchanged and still the only record when this is off. ⚠️ This can write secrets into your logs: shell commands routinely carry credentials (export TOKEN=…, curl -H 'Authorization: Bearer …', ssh/psql invocations). It is opt-in for that reason — enable it only where the log sink is trusted, and scrub or rotate the log afterwards. Only the command body is dumped, never Authorization headers or unrelated tool arguments. Scoped to the allow-list below; not a general tool-argument dump.
- MCP_LOG_RAW_COMMAND_TOOLS: (Optional) Comma-separated tool names eligible for raw-command logging, overriding the default post_command_execute_shell,executeShell. Entries that are not valid tool names are ignored. Use this when a new terminal-style upstream lands.
Verified Clients & Live Results (2026-06-12)
The example configurations below were exercised against the live APIs, and the proxy was attached to a range of agent CLIs over stdio MCP. Results from that verification run:
API example results
| API example | Tools | Sample call proven | Extra env needed |
|---|---|---|---|
| glama | 6 | get_v1_attributes |
none |
| apis.guru | 7 | get_metrics_json |
none |
| wolframalpha | 2 | get_v1_llm_api |
API_KEY |
| virustotal | 4 | IP report | API_KEY + API_AUTH_TYPE=api-key + API_AUTH_HEADER=x-apikey |
| asana | 73 (whitelist /workspaces,/projects,/tasks) |
created project + 11 tasks | SERVER_URL_OVERRIDE + API_KEY |
| render | 52 | get_services |
API_KEY (NEW spec URL render-public-api-1.json) |
| notion | 4–5 (whitelist /v1/users,/v1/search or /v1/pages) |
page create + title read-back | SERVER_URL_OVERRIDE + EXTRA_HEADERS (Notion-Version) + API_KEY |
| elevenlabs | 19 | TTS mp3 generated | SERVER_URL_OVERRIDE + API_AUTH_TYPE=api-key + API_AUTH_HEADER=xi-api-key |
| flyio | 34–35 | apps + machine health | API_KEY |
| slack | 7 (exact dot-path whitelist until #27 fix) | auth.test + postMessage |
API_KEY |
| netbox | 9 (whitelist /ipam/ip-addresses) |
IPAM write + read | API_KEY + API_AUTH_TYPE=Token |
| homeassistant | 21 | get_config (200) + call_service (light.turn_on, 200) |
SERVER_URL_OVERRIDE + API_KEY (${HA_TOKEN}, sent as Bearer) |
Client matrix
| Agent CLI | Model used (live test) | MCP attach mechanism | Tool calls | Prompts/Resources surfaced to model? |
|---|---|---|---|---|
| opencode | (CLI default) | ~/.config/opencode/opencode.json
mcp |
✅ native | prompts: ✅ (slash) · resources: ✅ — most complete ‖ |
| Codex | gpt-5-codex (OpenAI API) |
codex exec -c mcp_servers.* |
✅ native | prompts: ❌ low-level · ✅ fastmcp (via call_function→get_prompt) · resources: ✅ (read_mcp_resource) ‖ |
| Kilocode | kilo-auto/free |
global settings/mcp_settings.json |
✅ native | prompts: ❌ (no prompt mechanism) · resources: ✅ (access_mcp_resource) ‖ |
| Qwen | agent group via local LiteLLM gateway |
project .qwen/settings.json |
✅ native (live invoke auth-blocked) | prompts: ✅ (slash /summarize_spec) · resources: ❌ (no client support) ‖ |
| Gemini | Google OAuth free tier (CLI default model) | project .gemini/settings.json
mcpServers |
✅ native | prompts: interactive slash only · resources: interactive @ only (neither reaches the model headless) ‖ |
| Vibe | mistral-medium-3.5 |
~/.vibe/config.toml
[[mcp_servers]] |
✅ discovery + reads | prompts: ❌ · resources: ❌ (tools-only client) ‖ |
| agy | — | — | ❌ headless cannot enable MCP | n/a |
| letta (cloud / self-hosted) | Letta Cloud / PUT /v1/tools/mcp/servers |
streamable-HTTP / stdio | ✅ (per prior sweep) | unknown — needs a running Letta server to test ‡ |
> How to read the prompts/resources column. MCP has three surfaces — tools, prompts,
> resources. mcp-openapi-proxy serves all three, advertised by default since 0.3.0.
> Whether they reach the model is up to the client, and that varies:
>
> - ‖ re-verified 2026-06-15 against the published 0.3.3 release with no flags set
> (validating the default-on advertising), driving each real client binary, in
> both server modes (low-level and FastMCP simple). Per-client results were the
> same across modes except where noted.
> - Tools work on every client tested. Prompts→model: opencode & Qwen (slash
> commands); Gemini interactive-only. Resources→model: opencode, Codex, Kilocode.
> opencode is the only client that surfaces all three in any mode.
> - FastMCP simple mode exception: a client with no native prompt surface can still
> reach prompts through the static call_function→get_prompt indirection. Observed
> with Codex (prompts ❌ in low-level, ✅ in FastMCP simple mode). Vibe stays
> tools-only regardless of mode.
> - ‡ unknown — Letta can't be exercised without standing up a Letta server + model;
> left untested rather than guessed.
> - Every cell on the proxy side was confirmed via a raw stdio handshake (initialize
> advertises tools+prompts+resources; prompts/get and resources/read return content).
> The remaining ❌ are client-side gaps, not proxy limitations.
>
> (All pre-0.3.0 prompt/resource findings were voided — they were measured while
> advertising defaulted off, so they reflected the server default, not the clients.)
Minimal sanitized configs per client (the no-auth Glama spec is used as the smallest working example; substitute your own spec URL and $YOUR_KEY as needed):
Codex — inline -c mcp_servers.* overrides```bash
codex exec \
-c 'mcp_servers.glama.command="uvx"' \
-c 'mcp_servers.glama.args=["mcp-openapi-proxy"]' \
-c 'mcp_servers.glama.env.OPENAPI_SPEC_URL="https://glama.ai/api/mcp/openapi.json"' \
"list the available tools"
Gemini — project .gemini/settings.json```json
{
"mcpServers": {
"glama": {
"command": "uvx",
"args": ["mcp-openapi-proxy"],
"env": { "OPENAPI_SPEC_URL": "https://glama.ai/api/mcp/openapi.json" }
}
}
}
Qwen — project .qwen/settings.json```json
{
"mcpServers": {
"glama": {
"command": "uvx",
"args": ["mcp-openapi-proxy"],
"env": { "OPENAPI_SPEC_URL": "https://glama.ai/api/mcp/openapi.json" }
}
}
}
Kilocode — global settings/mcp_settings.json (use a clean workspace)```json
{
"mcpServers": {
"glama": {
"command": "uvx",
"args": ["mcp-openapi-proxy"],
"env": { "OPENAPI_SPEC_URL": "https://glama.ai/api/mcp/openapi.json" }
}
}
}
opencode — ~/.config/opencode/opencode.json mcp block```json
{
"mcp": {
"glama": {
"type": "local",
"command": ["uvx", "mcp-openapi-proxy"],
"environment": { "OPENAPI_SPEC_URL": "https://glama.ai/api/mcp/openapi.json" }
}
}
}
Vibe — ~/.vibe/config.toml [[mcp_servers]]```toml
[[mcp_servers]]
name = "glama"
command = "uvx"
args = ["mcp-openapi-proxy"]
[mcp_servers.env]
OPENAPI_SPEC_URL = "https://glama.ai/api/mcp/openapi.json"
agy (headless) and letta cloud could not attach a stdio MCP server at all, so no config snippet applies: agy provides no way to enable MCP in headless mode, and letta cloud only accepts remote MCP servers.
Tips
- Big specs need
TOOL_WHITELIST. Large APIs (Asana, Render, Fly.io, Slack, NetBox) expose dozens of endpoints; whitelist the paths you need to keep the tool count manageable for clients. - Dot-style paths need exact whitelist entries for now. Slack-style paths like
/chat.postMessageare not matched by prefix whitelisting — list each path exactly (issue #27 / PR #32). - Slow remote specs can crash-loop short-timeout clients. If a client kills the server before a large spec finishes downloading, fetch the spec once and point
OPENAPI_SPEC_URLat a localfile://copy (issue #28). - Custom auth schemes are handled via
API_AUTH_TYPE(PR #25), e.g.Tokenfor NetBox, orapi-keywithAPI_AUTH_HEADERfor Virustotal/ElevenLabs-style header keys.
Examples
For testing you can run the uvx command as demonstrated in the examples then interact with the MCP server via JSON-RPC messages to list tools and resources. See the "JSON-RPC Testing" section below.
Each example below is collapsed — click to expand.
Glama Example — the most minimal config: just OPENAPI_SPEC_URL, no auth
Glama offers the most minimal configuration for mcp-openapi-proxy requiring only the OPENAPI_SPEC_URL environment variable. This simplicity makes it ideal for quick testing.
1. Verify the OpenAPI Specification
Retrieve the Glama OpenAPI specification:```bash
curl https://glama.ai/api/mcp/openapi.json
Ensure the response is a valid OpenAPI JSON document.
#### 2. Configure mcp-openapi-proxy for Glama
Add the following configuration to your MCP ecosystem settings:```json
{
"mcpServers": {
"glama": {
"command": "uvx",
"args": ["mcp-openapi-proxy"],
"env": {
"OPENAPI_SPEC_URL": "https://glama.ai/api/mcp/openapi.json"
}
}
}
}
3. Testing
Start the service with:```bash
OPENAPI_SPEC_URL="https://glama.ai/api/mcp/openapi.json" uvx mcp-openapi-proxy
Then refer to the JSON-RPC Testing section for instructions on listing resources and tools.
Fly.io Example — machines API with an API token
Fly.io provides a simple API for managing machines making it an ideal starting point. Obtain an API token from Fly.io documentation.
#### 1. Verify the OpenAPI Specification
Retrieve the Fly.io OpenAPI specification:```bash
curl https://raw.githubusercontent.com/abhiaagarwal/peristera/refs/heads/main/fly-machines-gen/fixed_spec.json
Ensure the response is a valid OpenAPI JSON document.
2. Configure mcp-openapi-proxy for Fly.io
Update your MCP ecosystem configuration:```json
{
"mcpServers": {
"flyio": {
"command": "uvx",
"args": ["mcp-openapi-proxy"],
"env": {
"OPENAPI_SPEC_URL": "https://raw.githubusercontent.com/abhiaagarwal/peristera/refs/heads/main/fly-machines-gen/fixed_spec.json",
"API_KEY": "<your_flyio_token_here>"
}
}
}
}
- **OPENAPI_SPEC_URL**: Points to the Fly.io OpenAPI specification.
- **API_KEY**: Your Fly.io API token (replace ``).
- **API_AUTH_TYPE**: Set to `Api-Key` for Fly.io’s header-based authentication (overrides default `Bearer`).
#### 3. Testing
After starting the service refer to the JSON-RPC Testing section for instructions on listing resources and tools.
Render Example — manage hosted services with a tool whitelist
Render offers infrastructure hosting that can be managed via an API. The provided configuration file `examples/render-claude_desktop_config.json` demonstrates how to set up your MCP ecosystem quickly with minimal settings.
#### 1. Verify the OpenAPI Specification
Retrieve the Render OpenAPI specification:```bash
curl https://api-docs.render.com/openapi/render-public-api-1.json
Ensure the response is a valid OpenAPI document.
2. Configure mcp-openapi-proxy for Render
Add the following configuration to your MCP ecosystem settings:```json
{
"mcpServers": {
"render": {
"command": "uvx",
"args": ["mcp-openapi-proxy"],
"env": {
"OPENAPI_SPEC_URL": "https://api-docs.render.com/openapi/render-public-api-1.json",
"TOOL_WHITELIST": "/services,/maintenance",
"API_KEY": "your_render_token_here"
}
}
}
}
#### 3. Testing
Launch the proxy with your Render configuration:```bash
OPENAPI_SPEC_URL="https://api-docs.render.com/openapi/render-public-api-1.json" TOOL_WHITELIST="/services,/maintenance" API_KEY="your_render_token_here" uvx mcp-openapi-proxy
Then refer to the JSON-RPC Testing section for instructions on listing resources and tools.
Slack Example — payload token stripping via JMESPath (STRIP_PARAM)
Slack’s API showcases stripping unnecessary token payload using JMESPath. Obtain a bot token from Slack API documentation.
1. Verify the OpenAPI Specification
Retrieve the Slack OpenAPI specification:```bash
curl https://raw.githubusercontent.com/slackapi/slack-api-specs/master/web-api/slack_web_openapi_v2.json
Ensure it’s a valid OpenAPI JSON document.
#### 2. Configure mcp-openapi-proxy for Slack
Update your configuration:```json
{
"mcpServers": {
"slack": {
"command": "uvx",
"args": ["mcp-openapi-proxy"],
"env": {
"OPENAPI_SPEC_URL": "https://raw.githubusercontent.com/slackapi/slack-api-specs/master/web-api/slack_web_openapi_v2.json",
"TOOL_WHITELIST": "/chat,/bots,/conversations,/reminders,/files,/users",
"API_KEY": "<your_slack_bot_token, starts with xoxb>",
"STRIP_PARAM": "token",
"TOOL_NAME_PREFIX": "slack_"
}
}
}
}
- OPENAPI_SPEC_URL: Slack’s OpenAPI spec URL.
- TOOL_WHITELIST: Limits tools to useful endpoint groups (e.g. chat, conversations, users).
- API_KEY: Your Slack bot token (e.g.
xoxb-..., replace ``). - STRIP_PARAM: Removes the token field from the request payload.
- TOOL_NAME_PREFIX: Prepends
slack_to tool names.
3. Testing
After starting the service refer to the JSON-RPC Testing section for instructions on listing resources and tools.
GetZep Example — project-generated spec with Api-Key auth
> Note (2026): GetZep's hosted endpoint (api.getzep.com) returns 401 for previously-working keys — the hosted free tier no longer accepts API keys. The example now targets a self-hosted Zep CE: see examples/zep-ce/docker-compose.yml (zep built from source tag v1.0.2 + pgvector/pgvector:pg17; the delisted zepai/zep image must be rebuilt via the official Dockerfile.ce). Auth quirk: Zep's scheme is literally Api-Key, which collides with the proxy's special api-key mode — use API_AUTH_TYPE=api-key with API_KEY="Api-Key ". Live-verified: 8 tools via TOOL_WHITELIST=/sessions; session create + read-back through the bridge.
GetZep offers a free cloud API for memory management with detailed endpoints. Since GetZep did not provide an official OpenAPI specification, this project includes a generated spec hosted on GitHub for convenience. Users can similarly generate OpenAPI specs for any REST API and reference them locally (e.g. file:///path/to/spec.json). Obtain an API key from GetZep's documentation.
1. Verify the OpenAPI Specification
Retrieve the project-provided GetZep OpenAPI specification:```bash
curl https://raw.githubusercontent.com/matthewhand/mcp-openapi-proxy/refs/heads/main/examples/getzep.swagger.json
Ensure it’s a valid OpenAPI JSON document. Alternatively, generate your own spec and use a `file://` URL to reference a local file.
#### 2. Configure mcp-openapi-proxy for GetZep
Update your configuration:```json
{
"mcpServers": {
"getzep": {
"command": "uvx",
"args": ["mcp-openapi-proxy"],
"env": {
"OPENAPI_SPEC_URL": "https://raw.githubusercontent.com/matthewhand/mcp-openapi-proxy/refs/heads/main/examples/getzep.swagger.json",
"TOOL_WHITELIST": "/sessions",
"API_KEY": "<your_getzep_api_key>",
"API_AUTH_TYPE": "Api-Key",
"TOOL_NAME_PREFIX": "zep_"
}
}
}
}
- OPENAPI_SPEC_URL: Points to the project-provided GetZep Swagger spec (or use
file:///path/to/your/spec.jsonfor a local file). - TOOL_WHITELIST: Limits to
/sessionsendpoints. - API_KEY: Your GetZep API key.
- API_AUTH_TYPE: Uses
Api-Keyfor header-based authentication. - TOOL_NAME_PREFIX: Prepends
zep_to tool names.
3. Testing
After starting the service refer to the JSON-RPC Testing section for instructions on listing resources and tools.
Virustotal Example — YAML spec + custom x-apikey auth header
This example demonstrates:
- Using a YAML OpenAPI specification file
- Using custom HTTP auth header, "x-apikey"
1. Verify the OpenAPI Specification
Retrieve the Virustotal OpenAPI specification:```bash
curl https://raw.githubusercontent.com/matthewhand/mcp-openapi-proxy/refs/heads/main/examples/virustotal.openapi.yml
Ensure that the response is a valid OpenAPI YAML document.
#### 2. Configure mcp-openapi-proxy for Virustotal
Add the following configuration to your MCP ecosystem settings:```json
{
"mcpServers": {
"virustotal": {
"command": "uvx",
"args": ["mcp-openapi-proxy"],
"env": {
"OPENAPI_SPEC_URL": "https://raw.githubusercontent.com/matthewhand/mcp-openapi-proxy/refs/heads/main/examples/virustotal.openapi.yml",
"EXTRA_HEADERS": "x-apikey: ${VIRUSTOTAL_API_KEY}",
"OPENAPI_SPEC_FORMAT": "yaml"
}
}
}
}
Key configuration points:
- By default, the proxy expects a JSON specification and sends the API key with a Bearer prefix.
…(正文过长,页面只展示前面部分,完整内容见仓库)
