Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/1mcp-app/agent/servegit clone --depth 1 https://github.com/1mcp-app/agentWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00032 | $0.03090 |
| Opus 5 | $0.00016 | $0.01545 |
| Sonnet 5 | $0.00006 | $0.00618 |
| Haiku 4.5 | $0.00003 | $0.00309 |
Grade A, and why
serve scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 291 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Serve Command
1mcp serve starts the main 1MCP runtime.
It is the process that aggregates your configured MCP servers, exposes the HTTP MCP surface, initializes presets and instruction aggregation, and resolves template servers when client or session context becomes available.
Synopsis
1mcp serve [options]
1mcp [options]
serve is the default command.
When to Use serve
Use serve whenever you want to:
- run the aggregated 1MCP runtime
- power CLI mode for agents
- expose a direct HTTP MCP endpoint to MCP-native clients
- provide a runtime for
1mcp proxyto bridge stdio-compatible clients with project context
CLI mode depends on a running serve instance.
Current Mental Model
serve is not just a transport switch. It is the main runtime process.
- Static servers are created from startup configuration.
- Template servers are created later per client or session.
- Async loading can start HTTP availability before all servers finish loading.
- Lazy loading can keep server exposure narrower until needed.
- Instruction aggregation and preset notifications are initialized inside this runtime.
For runtime-wide configuration details, see the Configuration Guide.
Common Options
Configuration
--config, -c <path>: Specify a configuration file.--config-dir, -d <path>: Specify the config directory.
HTTP runtime
--port, -P <port>: Change the HTTP port. Default:3050.--host, -H <host>: Change the bind host. Default:localhost.--external-url <url>: Set the external base URL, usually for auth-related flows.
Filtering and presets
--filter, -f <expression>: Filter exposed servers with simple comma-separated tags or advanced boolean expressions.
Security
--enable-auth: Enable OAuth-backed auth on the runtime.--enable-enhanced-security: Enable additional security middleware.--trust-proxy <config>: Configure trusted reverse-proxy behavior.--template-context-trust <verified|disabled|legacy>: Control whether request context may render template servers. Default:verified.--confirm-untrusted-template-context: Required withlegacywhen binding HTTP to a non-loopback host.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 291 lines · 32 tokens per session scan A e484c6a4324c
serve is a command published in the GitHub repository 1mcp-app/agent (494 stars, last pushed 3d ago), licensed Apache-2.0. It adds 32 tokens to every session and 3,090 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
go
MCP tool management and workflow proxy.
strict-development-rules
These rules apply strictly to all code modifications and additions within the Gemini CLI project.
release-harn
Run the tag-first Harn release workflow.
harn-release
Run the tag-first Harn release workflow.
web-component
Create a new web component following the established project conventions.
send
Send a WhatsApp message using the wu MCP server.