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/pshkv/sint-protocol/claude-mdgit clone --depth 1 https://github.com/pshkv/sint-protocolWhat 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 | $0.02361 | $0.02361 |
| Opus 5 | $0.01180 | $0.01180 |
| Sonnet 5 | $0.00472 | $0.00472 |
| Haiku 4.5 | $0.00236 | $0.00236 |
Grade A, and why
sint-protocol CLAUDE.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 yesterday.
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 — 216 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SINT Protocol — Agent Guide
This file helps AI agents (and humans) understand and contribute to the SINT Protocol codebase.
What is this?
SINT is a security enforcement layer for physical AI. It sits between AI agents and the physical world (robots, tool calls, actuators) ensuring every action is authorized, constrained, and audited.
Quick Commands
pnpm install # Install dependencies
pnpm run build # Build all packages (required before test)
pnpm run test # Run all tests (currently 1,105 passing)
pnpm run typecheck # Type-check without emitting
pnpm run clean # Remove build artifacts
pnpm run bench # Run PolicyGateway performance benchmarks (p50/p99 latency)
Run a single package:
pnpm --filter @sint/gate-policy-gateway test
pnpm --filter @sint/bridge-mcp test
Start the gateway server:
pnpm --filter @sint/gateway-server dev
Monorepo Layout
apps/gateway-server/ → Hono HTTP API (port 3000)
packages/core/ → Shared types, Zod schemas, tier constants
packages/capability-tokens/ → Ed25519 token issuance, delegation, validation
packages/policy-gateway/ → THE choke point: tier assignment, constraints, combos
packages/evidence-ledger/ → SHA-256 hash-chained audit log
packages/bridge-mcp/ → MCP tool call → SINT request mapping
packages/bridge-ros2/ → ROS 2 topic/service → SINT request mapping
packages/persistence/ → Storage interfaces + in-memory implementations
packages/conformance-tests/ → Security regression suite (must pass on every PR)
Architecture Rules
1. Every action flows through PolicyGateway.intercept()
No bridge adapter, route handler, or service should make authorization decisions independently. All requests go through the gateway.
2. Result<T, E> — never throw
All fallible operations return { ok: true, value: T } | { ok: false, error: E }. Use the ok() and err() helpers from @sint/core. Never use try/catch for control flow.
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.
- yesterday First seen · 216 lines · 2,361 tokens per session scan A ea5322365723
sint-protocol CLAUDE.md is an instructions file published in the GitHub repository pshkv/sint-protocol (9 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 2,361 tokens to every session, about $0.0118 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
agent-leash AGENTS.md
Instructions for tonydzi/agent-leash, covering agents.md — working in this repo, what this repo is, layout, how to verify a change and conventions — the claim discipline.
Doberman-Core AGENTS.md
Instructions for DobermanCore/Doberman-Core, covering claude.md — doberman operating manual, 0. on startup (every session), 1. what this repository is, 2. architecture & extension points and 3. prime directives (non-negotiable).
openfirma AGENTS.md
AGENTS.md instructions for Firma-AI/openfirma, covering agents.md, key commands, formatting, linting rules and api stability.
ardur AGENTS.md
Instructions for ArdurAI/ardur, covering ardur agent instructions, 0. first action in every new session, 1. what ardur is, and what it does not claim, 2. tl;dr commands and 3. architecture and trust boundaries.
ardur CLAUDE.md
Instructions for ArdurAI/ardur: This repository keeps a single, canonical set of agent instructions in AGENTS.md. This file exists only so that Claude Code loads them automatically; it deliberately holds no rules of its own, so that the two can never drift apart.
ardur GEMINI.md
Instructions for ArdurAI/ardur: This repository keeps a single, canonical set of agent instructions in AGENTS.md. This file exists only so that the Gemini CLI loads them automatically; it deliberately holds no rules of its own, so that the two can never drift apart.