agentmods.dev

abp-mcp CLAUDE.md

abp-mcp CLAUDE.md is an instructions file for coding agents from tekthar/abp-mcp. It costs 1,511 tokens per session, scanned A, original, MIT.

Project-specific instructions for abp-mcp, a .NET package that exposes selected ABP Framework application services as tools for AI assistants.

In plain words
What is it for?
Use them when changing the package, its API-definition reader, tool registry, MCP endpoint, error mapping, or tests.
Why use it?
They explain the project’s architecture and non-negotiable design rules, helping prevent changes that bypass ABP metadata or break permission handling.

Instructions file

not rated 4 1mo ago A scan 1,511 tokens original MIT
View the source file tekthar/abp-mcp
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/tekthar/abp-mcp/claude-md
Clone the repo
git clone --depth 1 https://github.com/tekthar/abp-mcp

Wrote 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.

agentmods badge for abp-mcp CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/tekthar/abp-mcp/claude-md.svg)](https://agentmods.dev/instructions/tekthar/abp-mcp/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/tekthar/abp-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/tekthar/abp-mcp/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,511 This file is loaded in full into every session.
When invoked 1,511 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.

ModelPer sessionOnce invoked
Fable 5 $0.01511 $0.01511
Opus 5 $0.00756 $0.00756
Sonnet 5 $0.00302 $0.00302
Haiku 4.5 $0.00151 $0.00151

Measured 5d ago against content hash 535d2e8c6741, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

abp-mcp 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 5d 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.

CLAUDE.md · 100 lines

How it starts

The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

Project-specific guidance for Claude Code (and any agent) working on abp-mcp.

Project summary

abp-mcp is an open-source NuGet package that turns any ABP Framework application into a Model Context Protocol server. It reads ABP's /api/abp/api-definition metadata, filters by [McpTool]-tagged services, and exposes them as MCP tools with permission-aware visibility and ABP-native error mapping. Target framework: net9.0. Current ABP reference version: 9.2.*. Current MCP SDK: ModelContextProtocol.AspNetCore 1.0.

Architecture at a glance

Program.cs ──▶ AddAbpMcp + AbpMcpModule
                        │
                        ▼
        IApiDefinitionReader ──▶ IToolDescriptorBuilder
                        │                 │
                        ▼                 ▼
                 IDynamicMcpToolRegistry  (initialized at app start)
                        │
                        ▼
        /mcp endpoint (Microsoft SDK) ──▶ IAbpMcpDispatcher

Every piece has exactly one job; none of them owns more than ~200 LOC.

Non-negotiable design rules

  1. Source of truth is ABP's api-definition, not OpenAPI and not raw reflection. If a change would make us re-implement metadata extraction, stop and reconsider.
  2. Tool exposure is opt-in. A service without [McpTool] is invisible. Do not add "expose everything" modes without an explicit design change.
  3. Permission check before invoke. Permission filter before list. Both paths. Test both.
  4. No catch-all exception handlers in the dispatcher. Use the mapping table in AbpMcpDispatcher.MapException. Every ABP exception type has a specific translation. Unknown exceptions become INTERNAL with no stack trace leaked.
  5. In-process only for v0.1 and v1.0. No sidecar processes. No stdio transport. Streamable HTTP only.
  6. ABP-idiomatic public surface. Attribute names mirror [RemoteService] ergonomics ([McpTool]/[McpIgnore]). Options collection mirrors ConventionalControllers.Create(assembly, ...). Configure<AbpMcpOptions> works in PreConfigureServices. If a new option breaks this convention, find another way.
  7. /mcp/_discover and /mcp/_explain endpoints stay shipped. They are the diff between "auto-generator that frustrates devs" and "auto-generator that delights them." Do not move them behind a feature flag, do not lazy-load them, do not skimp on the explain reasons.

Read the full file on GitHub · 100 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.

  1. 5d ago First seen · 100 lines · 1,511 tokens per session scan A 535d2e8c6741

Subscribe to this mod's changes

abp-mcp CLAUDE.md is an instructions file published in the GitHub repository tekthar/abp-mcp (4 stars, last pushed 1mo ago), licensed MIT. It adds 1,511 tokens to every session, about $0.0076 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.

abpabp-frameworkai-agentsaspnetcoreclaudedotnet
Related

Other instructions, from other repositories

FamilyNido CLAUDE.md

Instructions for pablitofernandez/FamilyNido, covering claude.md, c# development section start, c# instructions, general instructions and naming conventions.

pablitofernandez/FamilyNido · 1,761 tokens

hangfire-mcp-dotnet CLAUDE.md

Claude Code instructions for NikiforovAll/hangfire-mcp-dotnet, covering claude.md, project, commands, or directly and architecture.

NikiforovAll/hangfire-mcp-dotnet · 969 tokens

ABP-ai-skills CLAUDE.md

Claude Code instructions for smss123/ABP-ai-skills, covering abp framework — development instructions for claude code, how to use these instructions, reference files, core abp principles (always apply) and typical layered solution structure.

smss123/ABP-ai-skills · 2,099 tokens

ABP-ai-skills copilot-instructions.md

Copilot instructions for smss123/ABP-ai-skills, covering abp framework — github copilot custom instructions, how to answer abp questions, core abp principles (always apply), typical layered solution structure and reusable agent prompts.

smss123/ABP-ai-skills · 1,629 tokens

maui ci-copilot-pipeline-security.instructions.md

Security rules for the Copilot PR-review pipeline. Read before editing.

dotnet/maui · 1,393 tokens

maui performance-hotpaths.instructions.md

Instructions for dotnet/maui, covering performance-critical path rules, hot paths in maui, allocation avoidance, caching and invalidation and collection iteration.

dotnet/maui · 441 tokens