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/crackish/mcp-creatio/agents-mdgit clone --depth 1 https://github.com/CRACKISH/mcp-creatioWrote 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/crackish/mcp-creatio/agents-md)<a href="https://agentmods.dev/instructions/crackish/mcp-creatio/agents-md"><img src="https://agentmods.dev/badge/instructions/crackish/mcp-creatio/agents-md.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 | $0.10420 | $0.10420 |
| Opus 5 | $0.05210 | $0.05210 |
| Sonnet 5 | $0.02084 | $0.02084 |
| Haiku 4.5 | $0.01042 | $0.01042 |
Grade A, and why
mcp-creatio AGENTS.md scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
`EE certificate key too weak` — fetch a token with `curl -k` instead, see delegated/gateway). How it starts
The opening of the file, as written. The whole thing — 532 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Guide: MCP Creatio Server
This document is for AI coding agents contributing to this repository. It explains project purpose, architecture, invariants, and safe extension patterns.
1. Purpose
An implementation of a Model Context Protocol (MCP) server that exposes Creatio CRM to MCP-compatible AI clients (Claude Desktop, ChatGPT Connectors, GitHub Copilot, etc.).
Primary goals:
- Provide stable tool surface for CRUD, schema discovery, business process execution.
- Enforce safe data operations (especially Activities ownership rules, date/time UTC handling).
- Offer prompts that guide LLMs to use Creatio correctly.
2. High-Level Architecture
src/
creatio/ ← Low-level Creatio API client & auth providers
contracts/ ← provider interfaces (CrudProvider, ProcessProvider, … — the "ports")
services/ ← provider impls; CRUD backends in odata/ + dataservice/, plus process, sys-settings, …
engines/ ← domain layer over the contracts (readonly guard + audit; see below)
auth/ ← auth providers (legacy / OAuth2 client-credentials / stateless Bearer / broker) + core contract (contracts/headers/identity/constants)
server/ ← MCP server + HTTP layer (bearer edge, broker OAuth server, handlers)
mcp/ ← MCP tool descriptors, prompts, filters builder
tool-preparer.ts ← ToolPreparer/ToolRegistrar contracts (env-gated tools)
creatio-rest.ts ← shared REST/sys-setting contracts + helpers for capability clients
dataforge/ ← DataForge capability: client + tool preparer
globalsearch/ ← Global Search capability: client + tool preparer
bearer/ ← stateless per-request credential edge: Bearer OR forwarded cookie session (delegated: RFC 9728 + fail-fast expiry; gateway: trust)
oauth/ ← broker mode: the MCP's own OAuth 2.1 AS (DCR + /authorize + /token, JWT + PKCE)
sessions/ ← per-process MCP session/transport lifecycle + per-user Creatio token store (broker only)
utils/ ← Reusable helpers (env, network, context)
types/ ← Shared TypeScript interfaces & DTO shapes
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 · 532 lines · 10,420 tokens per session scan A 1d473f135a4c
mcp-creatio AGENTS.md is an instructions file published in the GitHub repository CRACKISH/mcp-creatio (6 stars, last pushed 2mo ago), licensed MIT. It adds 10,420 tokens to every session, about $0.0521 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
nextcrm-app AGENTS.md
Instructions for pdovhomilja/nextcrm-app, covering agents.md, 1. context-mode, tool hierarchy, forbidden actions and trigger commands.
scout AGENTS.md
Instructions for agno-agi/scout, covering agents.md, project overview, architecture, contextprovider and one write surface.
relaticle CLAUDE.md
Claude Code instructions for relaticle/relaticle, covering architecture, module boundaries (enforced by tests/arch/archtest.php), actions (the write path), i18n enforcement and chat.
macro AGENTS.md
AGENTS.md instructions for macro-inc/macro, a project described as: Macro is a unified workspace for teams: email, chat, docs, tasks, agents, calls, and CRM — @-linked together with shared AI memory.
DaybydayCRM AGENTS.md
AGENTS.md instructions for Bottelet/DaybydayCRM, covering daybydaycrm agent guide, documentation map, project snapshot, architecture rules and keep controllers thin.
munin CLAUDE.md
Instructions for getmunin/munin, covering munin — agent guide, repository layout, module map (packages/backend-core/src/modules/), integration architecture and mcp surface.