What is helio AGENTS.md?

Project-specific instructions for Helio, an open-source gateway that controls how AI agents use external tools and records what they do.

They are used when developing or reviewing Helio's proxy, dashboard, or software libraries.

They explain the project's multi-package structure, rules, and development commands so changes fit the existing system.

helio AGENTS.md — Instructions file by gethelio

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/gethelio/helio/agents-md

Clone the repo

git clone --depth 1 https://github.com/gethelio/helio

Made for: Codex, OpenCode.

Per session 5,481 This file is loaded in full into every session.

When invoked 5,481 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	$0.05481	$0.05481
Opus 5	$0.02740	$0.02740
Sonnet 5	$0.01096	$0.01096
Haiku 4.5	$0.00548	$0.00548

Measured 2d ago against content hash 0eeaa4716806, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

helio 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 2d 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.

AGENTS.md · 333 lines

How it starts

The opening of the file, as written. The whole thing — 333 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Helio

Helio is an open-source MCP governance proxy. It sits between AI agents and external tools, applying policies, enforcing limits, requiring evidence, routing approvals, and recording an audit trail without modifying the agent or the tools.

GitHub org: gethelio · npm scope: @gethelio · PyPI: helio · Config file: helio.yaml

Architecture

Hybrid model: MCP proxy (core) + thin client SDKs (Python first, TypeScript planned).

Agent → Helio Proxy (policies, audit, limits, approvals) → Upstream MCP Server → External Tools
              ↑ optional sideband
         Python SDK (evidence context, dependency annotations)

The proxy intercepts tools/call JSON-RPC methods. All other MCP methods pass through untouched.
The SDK communicates with the proxy via sideband HTTP. It NEVER makes governance decisions.
SDK constraint: must stay under 500 lines per language. Governance logic belongs in the proxy.

Monorepo Structure

pnpm workspaces. Three packages:

packages/proxy/         → @gethelio/proxy (Node.js, Hono, TypeScript)
packages/dashboard/     → @gethelio/dashboard (React 19 + Vite + Tailwind CSS, internal workspace package bundled into proxy dist; not published)
packages/python-sdk/    → helio on PyPI (Python, <500 lines)

Supporting directories: examples/, docs/, docker/, scripts/, .github/

Tech Stack

Runtime: Node.js 24+, TypeScript 5.x (strict mode), pnpm 11 workspaces
HTTP: Hono (@hono/node-server)
Validation: Zod v4 (single source of truth for helio.yaml schema)
YAML: js-yaml
Input matchers: custom dot-path ($.field) conditions with eq/neq/gt/gte/lt/lte/contains/regex operators — no JSONPath library
Globbing: picomatch (tool-name patterns); safe-regex2 (validates operator-supplied regex conditions to reject catastrophic backtracking)
Database: better-sqlite3 (SQLite)
File watching: chokidar
CLI: Commander.js
Dashboard: React 19, React Router 7, Vite, Tailwind CSS 4, Recharts 3
Slack: @slack/web-api
Testing: Vitest (JS), pytest (Python SDK)
Linting: ESLint + Prettier
Build: tsup (proxy), Vite (dashboard)
CI: GitHub Actions; Husky pre-commit hooks (secret scan + lint/format/typecheck + docs-drift check)

Read the full file on GitHub · 333 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.

2d ago First seen · 333 lines · 5,481 tokens per session scan A 0eeaa4716806

Subscribe to this mod's changes

helio AGENTS.md is an instructions file published in the GitHub repository gethelio/helio (10 stars, last pushed 3d ago), licensed Apache-2.0. It adds 5,481 tokens to every session, about $0.0274 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.

ai-agent mcp mcp-governance mcp-proxy model-context-protocol proxy

helio AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Helio

Architecture

Monorepo Structure

Tech Stack

What this file has done since we first saw it

Other instructions, from other repositories

mcp-uplift AGENTS.md

mcp-uplift CLAUDE.md

dynamic-mcp AGENTS.md

suse-ai-up AGENTS.md

dynamic-mcp CLAUDE.md

mcp-server-excel coverage-prevention-strategy.instructions.md