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/t4cceptor/centian/agents-mdgit clone --depth 1 https://github.com/T4cceptor/centianWhat 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 | $0.00951 | $0.00951 |
| Opus 5 | $0.00476 | $0.00476 |
| Sonnet 5 | $0.00190 | $0.00190 |
| Haiku 4.5 | $0.00095 | $0.00095 |
Grade A, and why
centian 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.
How it starts
The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines (Source of Truth)
This file is authoritative for repo workflows, structure, and expectations. Other agent guidance should link here and avoid duplication.
Repository Identity & Purpose
Centian is the MCP proxy for this repository.
- Repository owner:
T4cceptor - Repository name:
centian - Repository URL:
https://github.com/T4cceptor/centian - Primary role: proxy MCP communication through a single CLI-managed endpoint
- Core capabilities:
- Proxy MCP communication to downstream systems
- Log and monitor MCP requests and results
- Process requests and responses to inspect, modify, or block them
- Aggregate downstream systems behind one endpoint, using tool namespacing to keep them distinguishable
Project Structure & Module Organization
Centian is a Go CLI application. The executable entrypoint lives in cmd/main.go.
Basic repository structure:
cmd/: CLI entrypoint and top-level command wiringinternal/cli/: command handlers for config, logs, auth, processor, server, and completion flowsinternal/proxy/: MCP proxy server, downstream connection handling, tool-call context, and request/response processing hooksinternal/logging/: log readers and shared logging helpersinternal/config/: config models, validation, accessors, and config-oriented CLI operationsinternal/processor/: processor interfaces, scaffolding, and CLI processor execution supportinternal/config/: config models, validation, persistence, config CLI operations, and import of existing MCP config filesinternal/auth/: API key and authentication helpersinternal/common/: shared MCP models, auth context, utilities, and internal logging primitivesdocs/: architecture and setup documentationdemo/: demo assets, configs, and supporting scriptsscripts/: repository scriptstests/test_configs/: config fixtures and sample MCP client/server configstests/integrationtests/: end-to-end and processor integration coveragebuild/: compiled binaries and generated artifacts
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 · 55 lines · 951 tokens per session scan A 1a985b0d2c11
centian AGENTS.md is an instructions file published in the GitHub repository T4cceptor/centian (15 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 951 tokens to every session, about $0.0048 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
VaultysClaw CLAUDE.md
Instructions for vaultys/VaultysClaw, covering claude.md, commands, development, demo / simulator and build.
CORE CLAUDE.md
Instructions for DariuszNewecki/CORE, covering claude.md — core, what core is, and what you are, source layout, how to work in this repo and governed and prohibited surfaces.
orloj CLAUDE.md
Instructions for OrlojHQ/orloj, covering orloj claude instructions, critical repository rules and implementation notes.
sint-protocol AGENTS.md
Instructions for sint-ai/sint-protocol, covering sint protocol — ai agent guide, what this repo does, key invariants (never violate these), quick orientation and package dependency graph.
claude-skills CLAUDE.md
Instructions for mastepanoski/claude-skills, covering claude.md, repository purpose, testing skills locally, repository architecture and core structure.
ai-safe2-framework AGENTS.md
AGENTS.md instructions for CyberStrategyInstitute/ai-safe2-framework, covering ai safe² v3.1 agent entry point, start here, version model, normative interpretation and enforcement planes.