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 instructions/dheerax/opencode_mcp/agents-mdgit clone --depth 1 https://github.com/Dheerax/opencode_mcpWhat 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 | $0.01650 | $0.01650 |
| Opus 5 | $0.00825 | $0.00825 |
| Sonnet 5 | $0.00330 | $0.00330 |
| Haiku 4.5 | $0.00165 | $0.00165 |
Grade A, and why
opencode_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.
How it starts
The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
opencode-mcp-bridge — Agent Integration Guide
This file follows the AGENTS.md convention: a single, tool-agnostic set of instructions any coding agent (Claude Code, Antigravity, Cursor, Windsurf, Codex CLI, JetBrains AI, etc.) can read to work with this repository correctly.
Start here
Call the opencode_guide MCP tool first, before delegating anything. It returns the complete,
always-current behavioral guide (model selection, async workflow, batch orchestration, failure
diagnosis) generated live from the running bridge's actual state — treat it as authoritative over
the summary below if they ever disagree.
What this is
An MCP server (opencode-mcp-bridge) that lets any agent delegate coding, refactoring, testing, and
generation tasks to the OpenCode CLI as a subagent — via OpenCode's
own opencode serve HTTP/SSE API. One persistent server process, one shared event stream, no PTY, no
per-task re-authentication.
- Binary target:
opencode - Events log:
~/.opencode-mcp-events.jsonl - Bridge logs:
~/.opencode-bridge/logs/bridge.log - Checkpoints:
~/.opencode-bridge/checkpoints/ - Server singleton state:
~/.opencode-bridge/server.json
Known gotcha: provider auth
google/* models (Antigravity/Gemini) route through a third-party auth plugin that can intermittently
fail with ModelUnavailableError if its cached OAuth token has expired. If a session goes straight to
error, check opencode_health's last_error before assuming it's a bridge bug — that error means the
provider needs re-auth (opencode auth login <provider>, run manually in a terminal), not a code fix.
opencode/* free-tier models have no such dependency and always work. For image generation specifically,
consider a dedicated image-gen tool/provider if one is available in your environment — OpenCode's
google/* image models ride the same auth path and are equally exposed to the failure above.
If opencode_server_health reports the server unhealthy or wedged, opencode_server_restart forces a
clean respawn — but this is rarely needed manually, since the bridge already restarts automatically the
moment it detects a failed session.
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 · 132 lines · 1,650 tokens per session scan A 128ec3337191
opencode_mcp AGENTS.md is an instructions file published in the GitHub repository Dheerax/opencode_mcp (1 stars, last pushed 1mo ago), licensed MIT. It adds 1,650 tokens to every session, about $0.0083 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.
Other instructions, from other repositories
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
buildNext
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
langchain AGENTS.md
Instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.