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/initlabsai/vibekit/agents-mdgit clone --depth 1 https://github.com/initlabsai/vibekitWhat 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 | $0.03319 | $0.03319 |
| Opus 5 | $0.01659 | $0.01659 |
| Sonnet 5 | $0.00664 | $0.00664 |
| Haiku 4.5 | $0.00332 | $0.00332 |
Grade A, and why
vibekit 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 — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
The map of this repository, for people and coding agents alike. README.md
is the product page. Read docs/CONSTITUTION.md before structural changes.
VibeKit exposes Algorand capabilities through one tool contract across MCP, CLI, and an LLM agent loop, and ships an Explorer (terminal and web) that renders what those tools return.
What runs where
| Path | What it is |
|---|---|
packages/vibekit |
The one published package, @initlabs/vibekit. Every src/ area is a subpath export. |
apps/reference |
The reference agent: a Bun server mounting the toolkit and a Vite/React page over vibekit add components. Keep it small; it is the tutorial's endpoint. |
packages/vibekit/src/core (.) |
The tool contract, resolveDeployment, executeToolCall, the codec, and the compose engine (core/compose/). |
packages/vibekit/src/tools (./tools, ./tools/views) |
The domain tools: accounts, assets, contracts, network, transactions — each a flat directory of index.ts (the tool definitions), schemas.ts, and the functions behind them. views.ts maps each view id to its wire schema. |
packages/vibekit/src/plugins/* (./plugins/<name>) |
Third-party tool plugins: nfd, pera, vestige, alpha-arcade. |
packages/vibekit/src/signer-keystore (./signer-keystore) |
The algosdk signer over the keystore daemon, its account tools, the testnet dispenser. |
packages/vibekit/src/mcp (./mcp, ./mcp/stdio, ./mcp/http) |
Host: ToolDefinition-to-MCP adapter and transports. |
packages/vibekit/src/agent (./agent, ./agent/config) |
Host: the LLM tool loop and its stored config. |
packages/vibekit/src/preset (./preset) |
The stock tool and plugin mix every stock host composes from. |
packages/vibekit/examples |
Reference stdio and HTTP deployments, typechecked with the package. |
apps/cli |
The vibekit binary: new, init, localnet, keystore, dispenser, doctor, tool, mcp, explore. Host: commands/tool.ts and commands/mcp.ts. |
apps/tui |
The terminal Explorer (OpenTUI). Live against a network when reachable, sample data otherwise. features/<name>/ holds one feature's hooks, screen, and cards; feed/ is the transcript; app.tsx composes them. |
apps/agent |
The web Explorer with its agent (Next.js). Sample-backed reads plus a compose-only flow route. |
apps/website |
The public site (Astro/Starlight). |
skills/ |
Canonical skills, compiled into the CLI by bun run --cwd apps/cli bundle-skills. .agents/skills, .claude/skills, .grok/skills are symlinks into it. |
verify/ |
The packed-consumer gate (bun run verify:packed). The LocalNet smoke is apps/cli/scripts/smoke-localnet.ts (bun run smoke:localnet, needs vibekit localnet start; CI runs it). |
test-prompts/ |
Agent-run MCP acceptance prompts and their transcripts. |
ideas/ |
Dated design notes that are not yet work. |
out/ |
Marketing video build output (see the marketing-content skill). |
docs/ |
CONSTITUTION.md and PRODUCTION.md only. |
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 · 214 lines · 3,319 tokens per session scan A 16a3caccc66f
vibekit AGENTS.md is an instructions file published in the GitHub repository initlabsai/vibekit (27 stars, last pushed 3d ago), licensed MIT. It adds 3,319 tokens to every session, about $0.0166 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 instructions, from other repositories
algorand-mcp AGENTS.md
Instructions for GoPlausible/algorand-mcp, covering algorand mcp — agent guide, calling mcp tools, session start, network selection and amounts and decimals.
pumpfun-bonkfun-bot CLAUDE.md
Instructions for chainstacklabs/pumpfun-bonkfun-bot, covering agent guide, ground rules, layout, naming inside learning-examples/ and commands.
evmole AGENTS.md
Instructions for cdump/evmole: EVMole extracts structured facts from deployed EVM runtime bytecode.
VaultysClaw CLAUDE.md
Instructions for vaultys/VaultysClaw, covering claude.md, commands, development, demo / simulator and build.
plamen CLAUDE.md
Instructions for PlamenTSV/plamen, covering plamen — security auditor, execution model, reference files and shared.
solana-developer-platform AGENTS.md
Instructions for solana-foundation/solana-developer-platform, covering solana developer platform, repo layout, source of truth, generated files and public vs internal surfaces.