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 rules/gtorreal/buda-mcp/tool-output-formatgit clone --depth 1 https://github.com/gtorreal/buda-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/rules/gtorreal/buda-mcp/tool-output-format)<a href="https://agentmods.dev/rules/gtorreal/buda-mcp/tool-output-format"><img src="https://agentmods.dev/badge/rules/gtorreal/buda-mcp/tool-output-format.svg" alt="Measured on agentmods" 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 | $0.00000 | $0.00755 |
| Opus 5 | $0.00000 | $0.00378 |
| Sonnet 5 | $0.00000 | $0.00151 |
| Haiku 4.5 | $0.00000 | $0.00076 |
Grade A, and why
tool-output-format 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tool Output Format
Mandate
Every tool must return pre-formatted markdown, never raw JSON. This ensures any MCP client (Claude Desktop, Cursor, Cline, Zed, etc.) renders the same structured output regardless of AI-side configuration.
// NEVER do this
return { content: [{ type: "text", text: JSON.stringify(result, null, 2) }] };
// ALWAYS do this
return { content: [{ type: "text", text: formatMyTool(result) }] };
Shared helpers — src/format.ts
Import from ../format.js (compiled path). Available helpers:
| Helper | Purpose |
|---|---|
liquidityBadge(spreadPct) |
Returns ✅ / 🟡 / 🔴 (thresholds: <0.5%, <2%, ≥2%) |
fmtSlippage(pct | null) |
+0.052% / -0.008% / — for null |
fmtTimestamp(iso) |
"2026-04-13 15:10:33 UTC" from ISO string |
Add new helpers to src/format.ts when an existing one doesn't cover the use case. Keep helpers generic and reusable — not tied to a specific tool.
Pattern for every tool file
1. Import helpers
import { liquidityBadge, fmtSlippage, fmtTimestamp } from "../format.js";
2. Define a formatX() function in the tool file
The function takes the typed result object and returns a markdown string. Place it between the type definitions and handleX().
export function formatMyTool(result: MyToolResult): string {
return [
`## My Tool — ${result.market_id}`,
`_Updated: ${fmtTimestamp(result.timestamp)}_`,
"",
`Bid: ${result.bid} | Ask: ${result.ask}`,
].join("\n");
}
3. Use it in the handler return
return {
content: [{ type: "text", text: formatMyTool(result) }],
};
4. Export it for unit testing (optional but recommended)
Exporting formatX lets test/unit.ts assert on the rendered output without network calls.
Real example — src/tools/stable_liquidity.ts
formatStableLiquidity(results) produces:
## Stablecoin Liquidity — Buda.com
_Updated: 2026-04-13 15:10:33 UTC_
### USDC-CLP — Spread 0.19% ✅
Bid: 899.37 | Ask: 901.12 CLP
| Size | Buy slippage | Sell slippage |
|--------|-------------|---------------|
| $1k | +0.002% | -0.008% |
| $100k | +0.002% | -0.008% |
---
### USDT-COP — Spread 2.61% 🔴
...
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 · 93 lines · 0 tokens per session scan A a5ac57e53ebd
tool-output-format is a cursor rule published in the GitHub repository gtorreal/buda-mcp (0 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 755 tokens. 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 cursor rules, from other repositories
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.
control-plane-descriptors
Control plane descriptor and instance implementation patterns.
family-instance-domain-actions
Family instance domain action implementation patterns.