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/sevalla-hosting/mcp/agents-mdgit clone --depth 1 https://github.com/sevalla-hosting/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 | $0.01744 | $0.01744 |
| Opus 5 | $0.00872 | $0.00872 |
| Sonnet 5 | $0.00349 | $0.00349 |
| Haiku 4.5 | $0.00174 | $0.00174 |
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 3d ago.
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 — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for AI coding agents working in this repository.
Project Overview
Remote MCP server for the Sevalla PaaS API (mcp.sevalla.com). Exposes ~200 API endpoints
through 2 tools (search + execute) using sandboxed V8 isolates. Built with Hono,
MCP SDK, and isolated-vm. No build step — TypeScript runs natively on Node.js 24+.
Commands
pnpm dev # Development with --watch (native Node.js)
pnpm start # Run server (node src/index.ts)
pnpm test # Run all tests (node:test)
pnpm lint # oxlint
pnpm fmt # oxfmt (auto-fix)
pnpm fmt:check # oxfmt (check only)
pnpm check:code # Full check: tsc --noEmit --skipLibCheck && oxlint && oxfmt --check
Running a Single Test
node --test test/oauth.test.ts # Run one test file
node --test --test-name-pattern="encrypt" test/oauth.test.ts # Run matching tests
The test runner is node:test (built-in). No Jest, Vitest, or Mocha.
Type Checking
npx tsc --noEmit --skipLibCheck
Node.js Version
Node.js 24 required. See .nvmrc. The isolated-vm native addon segfaults on Node 25.
TypeScript runs via Node's built-in type stripping — there is no build/compile step.
Architecture
src/
├── index.ts # Main server: Hono app, MCP handler, graceful shutdown
├── oauth.ts # Stateless OAuth flow (HMAC-signed params, AES-256-GCM encrypted codes)
├── html.ts # Landing page HTML template
└── sandbox/
├── index.ts # Re-exports from all sandbox modules
├── bridge.ts # HTTP request bridge with security filters (path validation, header filtering)
├── isolate.ts # V8 isolate sandbox execution (isolated-vm)
├── spec.ts # OpenAPI spec processing ($ref resolution, tag extraction)
└── tools.ts # MCP tool definitions (search + execute)
Each HTTP POST to /mcp creates a fresh McpServer + CodeMode + transport bound to the
caller's API key. Fully stateless — no sessions, no server-side storage.
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.
- 3d ago First seen · 181 lines · 1,744 tokens per session scan A 4b14de1142b4
mcp AGENTS.md is an instructions file published in the GitHub repository sevalla-hosting/mcp (8 stars, last pushed 8d ago), licensed MIT. It adds 1,744 tokens to every session, about $0.0087 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
mcp CLAUDE.md
Instructions for shipstatic/mcp, covering claude.md, readme positioning (public docs), architecture, one product, two transports — the shape that survives oauth and quick reference.
mcp AGENTS.md
Instructions for shipstatic/mcp: Read CLAUDE.md (same directory). It is the single source of truth for this repo's instructions; everything in it applies to every agent, not just Claude. Do not duplicate its content here.
intervals-icu-mcp CLAUDE.md
Instructions for hhopke/intervals-icu-mcp, covering claude.md, project overview, development commands, architecture (quick reference) and tool categories.
ai-toolkit AGENTS.md
Instructions for pipefy/ai-toolkit, covering repository guidelines, documentation map, project structure, import namespace migration: pipefysdk → pipefy and src/pipefysdk/init.py (transitional shim).
flyto-core CLAUDE.md
Instructions for flytohub/flyto-core, covering claude notes, cross-agent handoff and shared code intelligence.
Plonk AGENTS.md
Instructions for ostapondo/Plonk, covering agent rules, layout, adding a module, build & verify and code style.