opencode-cli-mcp AGENTS.md

Repository instructions for sandraschi/opencode-cli-mcp, a project that lets AI agents run and manage the opencode command-line tool through an MCP server, with a web dashboard. MCP is a standard way for AI agents to call external tools.

In plain words
What is it for?
Use them when running, testing, checking, or modifying this project and its REST API, MCP endpoint, or dashboard.
Why use it?
They provide the project’s commands, ports, architecture, and coding standards so agents can work on the backend, frontend, and server consistently.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/sandraschi/opencode-cli-mcp/agents-md
Clone the repo
git clone --depth 1 https://github.com/sandraschi/opencode-cli-mcp

Made for: Codex, OpenCode.

Per session 654 This file is loaded in full into every session.
When invoked 654 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.00654 $0.00654
Opus 5 $0.00327 $0.00327
Sonnet 5 $0.00131 $0.00131
Haiku 4.5 $0.00065 $0.00065

Measured yesterday against content hash 9e2f18997f2e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

opencode-cli-mcp AGENTS.md 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.

AGENTS.md · 50 lines

How it starts

The opening of the file, as written. The whole thing — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.

opencode-cli-mcp - Agent Guide

Overview

MCP server wrapping the opencode CLI (opencode serve) for AI agent orchestration: launch/poll/review agent runs, session CRUD + diff, offline session depot over the opencode SQLite DB, .mcpb installs, and graceful shutdown. Ships a unified FastAPI backend (REST /api/* + FastMCP /mcp on one port) and a Vite/React fleet-standard dashboard.

Quick-Ref Commands

Command What it does
just serve Run MCP server (stdio)
just api Run unified backend (api.main, :10951)
just web Run Vite frontend (:10950)
just certify All gates: ruff, pytest, pyright, tsc, biome
just test / just check pytest / ruff check + format
.\start.ps1 Full stack: opencode serve + backend + frontend

Ports

Port Service
10950 Frontend (Vite)
10951 Unified backend (REST /api/* + MCP /mcp)
4096 opencode serve

Standards

  • FastMCP 3.2+ portmanteau tool pattern - tools use operation enum param; responses are structured dicts with success, message, domain-specific fields.
  • 6 primary tools (opencode_runs, opencode_sessions, opencode_depot, opencode_system, opencode_mcpb_install, opencode_shutdown) + 15 legacy atomic aliases (0.3.0 removes them).
  • Dual transport: stdio (Claude Desktop) + unified HTTP backend (api.main).
  • See mcp-central-docs for fleet-wide standards.

Key Files

File Purpose
src/opencode_cli_mcp/server.py FastMCP app + http_app (CORS-wrapped ASGI)
src/opencode_cli_mcp/client.py httpx client → opencode serve; OPENCODE_BINARY resolution
src/opencode_cli_mcp/depot.py Session depot (direct SQLite on opencode.db)
src/opencode_cli_mcp/registry.py Shared tool definitions
api/main.py Unified ASGI app (REST + MCP /mcp)
api/routes/capabilities.py /api/health, /api/v1/health, /api/v1/diagnostics
web_sota/src/store.ts Zustand store (incl. shared llmProvider/llmModel)
run_server.py PyInstaller entry point (OPENCODE_CLI_MCP_PORT → HTTP)

Read the full file on GitHub · 50 lines

Changes

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.

  1. yesterday First seen · 50 lines · 654 tokens per session scan A 9e2f18997f2e

Subscribe to this mod's changes

opencode-cli-mcp AGENTS.md is an instructions file published in the GitHub repository sandraschi/opencode-cli-mcp (2 stars, last pushed 1mo ago), licensed MIT. It adds 654 tokens to every session, about $0.0033 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 2026-08-31.