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/ditrixnew/mcp-db-client/copilot-instructionsgit clone --depth 1 https://github.com/DitriXNew/MCP-DB-ClientWhat 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 | $0.01010 | $0.01010 |
| Opus 5 | $0.00505 | $0.00505 |
| Sonnet 5 | $0.00202 | $0.00202 |
| Haiku 4.5 | $0.00101 | $0.00101 |
Grade A, and why
MCP-DB-Client copilot-instructions.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 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.
http1c — MCP Server Framework for 1C:Enterprise
Project Overview
This is a framework for building Model Context Protocol (MCP) servers from 1C:Enterprise. It consists of two parts:
- http-1c-dll/ — C++17 native DLL (1C Native API add-in) that handles MCP transport
- http-1c-dp/ — Reference 1C data processor (BSL module) showing how to implement tools, resources, and prompts
The DLL handles protocol-level concerns (HTTP, JSON-RPC, sessions, security). The BSL module handles business logic (tool handlers, resource data, prompt generation).
Architecture
MCP Client ←→ Native DLL (C++ httplib) ←→ 1C:Enterprise (BSL via ExternalEvent)
- POST /mcp — JSON-RPC MCP requests
- GET /mcp — SSE notification stream
- DELETE /mcp — Session termination
Key Conventions
C++ (http-1c-dll/src/)
- Build: CMake + MSVC, NMake Makefiles,
/MTstatic CRT,/utf-8 - JSON: nlohmann/json (
json.hpp) - HTTP: cpp-httplib (
httplib.h) - 1C API: Native API headers in
include/(AddInDefBase.h,IMemoryManager.h,types.h,ComponentBase.h) - String conversion:
MB2WCHAR()/WCHAR2MB()for UTF-8 ↔ UTF-16 - Thread safety: Separate mutexes per resource (
toolsMutex,resourcesMutex,sessionMutex, etc.) - Events to 1C:
ExtEvent(u"HttpServer", u"EventName", data) - Method registration:
AddProcedure()/AddFunction()/AddProperty()with EN/RU names - Version: Defined in
version.hasVERSION_SEMVER
BSL (http-1c-dp/http1c/Forms/Form/Ext/Form/Module.bsl)
- Client-side handlers are
&AtClientprocedures (no Async/Await) - Server calls for metadata/computation use
&AtServerfunctions - Component configuration via synchronous property assignments:
Component.PropertyName = Value - Component methods via
BeginCalling<MethodName>withNotifyDescriptioncallbacks - DO NOT use reserved 1C names as local variables (
Catalogs,Documents,Metadata, etc.) - JSON:
JSONReader/JSONWriter+ReadJSON()/WriteJSON() - Error pattern: Try/Except →
SendToolError()/SendResourceError()/SendPromptError() - Tool definitions:
NewTool(),AddToolParam(),AddToolAnnotations(),SetToolOutputSchema() - Resource definitions: Structure with
uri,name,description,mimeType - Prompt definitions: Structure with
name,description,argumentsarray
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 · 1,010 tokens per session scan A 068618270122
MCP-DB-Client copilot-instructions.md is an instructions file published in the GitHub repository DitriXNew/MCP-DB-Client (36 stars, last pushed 2mo ago), licensed MIT. It adds 1,010 tokens to every session, about $0.0051 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
EDT-MCP CLAUDE.md
Claude Code instructions for DitriXNew/EDT-MCP, covering edt-mcp — code conduct (the minefield map), ❌ hard don'ts (violating these = a bug or corruption), 🛑 "stop and think twice" zones (large blast radius), 🌐 two languages (ru/en) — the main recurring mine and ✅ before you write code.
EDT-MCP copilot-instructions.md
Copilot instructions for DitriXNew/EDT-MCP: Before making any significant changes, ALWAYS use expert consultation tools.
mcp-bsl-platform-help-context CLAUDE.md
Instructions for Desko77/mcp-bsl-platform-help-context, covering claude.md, project overview, commands, run the server — yaml config (recommended) and run the server — cli options (env vars mcpbsl also work).
intervals-icu-mcp CLAUDE.md
Instructions for hhopke/intervals-icu-mcp, covering claude.md, project overview, development commands, architecture (quick reference) and tool categories.
flyto-core CLAUDE.md
Instructions for flytohub/flyto-core, covering claude notes, cross-agent handoff and shared code intelligence.
rosetta AGENTS.md
Instructions for tikoci/rosetta, covering codex instructions for rosetta, first reads, development defaults and mcp and client config.