Borrowing it
Nothing to install: this file belongs to diblaze/bokio-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/diblaze/bokio-mcp/main/AGENTS.mdgit clone --depth 1 https://github.com/diblaze/bokio-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/diblaze/bokio-mcp/agents-md)<a href="https://agentmods.dev/instructions/diblaze/bokio-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/diblaze/bokio-mcp/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/diblaze/bokio-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/diblaze/bokio-mcp/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 5.1 | $0.03101 | $0.03101 |
| Opus 5.5 | $0.01240 | $0.01240 |
| Sonnet 5.5 | $0.00620 | $0.00620 |
| Haiku 4.5 | $0.00310 | $0.00310 |
Grade A, and why
bokio-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 9d 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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
bokio-mcp
Purpose
A Model Context Protocol server (stdio transport) that wraps the Bokio accounting REST API as MCP tools. Single-company, private Bearer-token auth. Read-only by default — all mutating tools are hidden and blocked unless BOKIO_ALLOW_WRITES=1. For small companies and consultants who keep their own books in Bokio; installed by users with npx -y bokio-mcp. See README.md for user-facing install/config.
Key Files
| File | Description |
|---|---|
src/index.ts |
Server entry: builds the McpServer with the INSTRUCTIONS sent at initialize (keep them short and true to the tools), registers all 60 tools via the tool() helper, gates main() on being the entry point. Exports server for tests. |
src/sie.ts |
SIE 4 export parser (PC8/CP437, #RAR 0, annulment pairs) and the pure öre computations behind the computed tools. |
src/bokio.ts |
HTTP client + config: bokioRequest (JSON, enforces the write gate), bokioDownload (binary GET), fileForm, filterQuery, resolveCompanyId, writesAllowed, downloadTarget, cached/clearCache (5-minute read cache, cleared by any write). |
test/bokio.test.ts |
Unit tests (mocked fetch) for the client, write gate, filterQuery, downloadTarget. |
test/server.test.ts |
In-memory Client↔Server test: asserts all tools register with explicit annotations; computed tools against a mocked fiscal-year listing + SIE download; TDQS lint (mcp-tdqs) over all tools. |
test/sie.test.ts |
Parser and computation tests on synthetic SIE built in the file. |
scripts/smoke.mjs |
Spawns built dist/index.js directly and through a symlink (as npm installs bins), runs the MCP handshake + tools/list, asserts read-only surface (writes hidden, all annotated). Run via pnpm run smoke. |
scripts/pack-check.mjs |
npm pack --dry-run: tarball holds only dist/, README, LICENSE, package.json. Run via pnpm run pack:check (CI). |
server.json |
MCP Registry metadata (io.github.diblaze/bokio-mcp, matches mcpName in package.json). release-please bumps both version fields; not in the tarball. |
biome.json |
Biome config: 4-space, 100 cols, double quotes, semicolons, organize-imports. |
pnpm-workspace.yaml |
Carries allowBuilds: esbuild: true (pnpm 11 needs it for vitest), lefthook: false (hooks come from prepare). |
LLM.md |
Setup guide addressed to an AI agent; linked from the README, not shipped in the tarball. |
lefthook.yml |
Git hooks, installed by pnpm install (prepare: lefthook install --reset-hooks-path, which also clears an old core.hooksPath): pre-commit Biome on staged files + actionlint on staged workflows (skipped when not on PATH), commit-msg commitlint (Conventional Commits, built-in config). |
.github/workflows/release-please.yml |
On push to main: release-please opens or updates the release PR; when a release is created, the publish job runs npm stage publish (OIDC, Node 24). Needs the RELEASE_PLEASE_TOKEN secret. |
release-please-config.json, .release-please-manifest.json |
release-please config (node, bump-minor-pre-major, v-prefixed tags, changelog sections Added/Fixed/Changed) and the last released version. |
.github/workflows/mcp-registry.yml |
Runs on each published release (and manual dispatch on main), gated by the mcp-registry environment's required reviewer: installs pinned mcp-publisher (SHA-256 checked), logs in with GitHub OIDC and publishes server.json to the MCP Registry. |
.github/workflows/ci.yml |
CI: Node matrix [22, 24], pnpm, commitlint over the PR's commits → actionlint (+ shellcheck) → biome ci → typecheck → test → build → smoke → pack:check. |
.github/CODEOWNERS |
* @diblaze — required reviewer on PRs. |
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.
- 9d ago Changed · +63 tokens per session fe1b2f787098
- 12d ago Changed · +2 lines · +235 tokens per session 97043b451164
- 14d ago Changed · +1 lines · +153 tokens per session b4462bc887a3
- 18d ago Changed · +20 lines · +1,149 tokens per session 7426c4434ea5
- 2mo ago First seen · 81 lines · 1,501 tokens per session scan A d84fde1f8492
bokio-mcp AGENTS.md is an instructions file published in the GitHub repository diblaze/bokio-mcp (2 stars, last pushed 9d ago), licensed MIT. It adds 3,101 tokens to every session, about $0.0124 per session on Opus 5.5. 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.
bokio-mcp AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| bokio-mcp AGENTS.mddiblaze/bokio-mcp | Instructions file | 2 | 9d ago | MIT | A | 3,101 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,649 | GitHub Copilot |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | today | MIT | A | 4,680 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,093 | GitHub Copilot |
| vscode model-prompts.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 2,022 | GitHub Copilot |