What is mcp AGENTS.md?

A project instruction file for browserview/mcp, an MCP server that lets coding agents use browserview.io. It documents commands, layout, conventions, testing, and release steps.

Use it when modifying or releasing browserview/mcp, especially its TypeScript server, API connection, streaming code, or registered tools.

It gives agents and developers shared project rules, reducing guesswork about how to build, run, test, and publish the server.

mcp AGENTS.md — Instructions file by browserview

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/browserview/mcp/agents-md

Clone the repo

git clone --depth 1 https://github.com/browserview/mcp

Made for: Codex, OpenCode.

Per session 585 This file is loaded in full into every session.

When invoked 585 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.

Model	Per session	Once invoked
Fable 5	$0.00585	$0.00585
Opus 5	$0.00293	$0.00293
Sonnet 5	$0.00117	$0.00117
Haiku 4.5	$0.00059	$0.00059

Measured yesterday against content hash cf708af2badd, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 · 30 lines

How it starts

The opening of the file, as written. The whole thing — 30 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md — @browserview/mcp

Model Context Protocol server for browserview.io (stdio transport). Standalone: it does NOT depend on the TypeScript SDK — the HTTP client is reimplemented inline.

Commands

Build: npm run build
Run locally: BROWSERVIEW_API_KEY=... node dist/index.js (exits 1 with help if the key is unset)
Test: none automated — smoke by running without a key (help + exit 1) and with a key via an MCP client.
Publish: npm publish (prepublishOnly rebuilds; .npmrc reads ${NPM_TOKEN} — kept in the gitignored .env: set -a; . ./.env; set +a; npm publish)

Layout

src/index.ts — everything: env handling, inline API client with 429/503 retries, an SSE consumer for the hosted agent (apiStream/parseSse/turnStream/collectTurn; POST streams never retried, mid-turn drops and 202s followed via /agent/events?after=&turn=&follow=1), and twelve registerTool calls (create_session, list_sessions, get_session, destroy_session, mint_session_token, get_session_replay, run_agent_task, agent_send_message, agent_reply, agent_interrupt, get_agent_conversation, list_agent_models).
dist/ — build output, gitignored, published via files: ["dist"]; bin is browserview-mcp.

Conventions

Env: BROWSERVIEW_API_KEY (required), BROWSERVIEW_BASE_URL (optional; legacy BROWSERVIEW_API_URL still honored with a stderr warning).
Version lives in both package.json and the McpServer({ version }) call — bump together.
Tool results are JSON strings aimed at agents; keep descriptions accurate, they are the agent-facing docs. Keep the README's tool table in sync with the inputSchemas.
Public surface only: never expose admin-only API endpoints/fields as tools. The production API (https://sessions.browserview.io) is the contract.

Release

Bump version in package.json and src/index.ts (McpServer).
npm run build, then a quick stdio smoke (node dist/index.js without a key should print help and exit 1).
Commit, tag v<version>, push with tags (remote: [email protected]:browserview/mcp.git).
set -a; . ./.env; set +a; npm publish

Read the full file on GitHub · 30 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.

yesterday First seen · 30 lines · 585 tokens per session scan A cf708af2badd

Subscribe to this mod's changes

mcp AGENTS.md is an instructions file published in the GitHub repository browserview/mcp (0 stars, last pushed 4d ago), licensed MIT. It adds 585 tokens to every session, about $0.0029 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.

Backend and APIs Git and workflow Testing

spec-kit AGENTS.md

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.

github/spec-kit · 7,104 tokens

vscode buildNext.instructions.md

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).

microsoft/vscode · 6,785 tokens

mcp AGENTS.md