What is ARR_MCP AGENTS.md?

Engineering instructions for ARR_MCP, a repository that connects services to AI clients through MCP, a standard way for applications to expose tools. They require test-driven development, API research, safe logging, conventional commit messages, and focused branches.

Use them when adding or changing a service or tool, researching an external API, writing tests, preparing commits, or releasing a feature in ARR_MCP.

They prevent common integration failures, such as coding against an incorrect API or corrupting the JSON-RPC stream by writing logs to standard output. TDD means writing and seeing a test fail before implementing the code.

How GauranshMathur/ARR_MCP configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to GauranshMathur/ARR_MCP. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/GauranshMathur/ARR_MCP/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/GauranshMathur/ARR_MCP

Made for: Codex, OpenCode.

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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/gauranshmathur/arr_mcp/agents-md/github.svg)](https://agentmods.dev/instructions/gauranshmathur/arr_mcp/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/gauranshmathur/arr_mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/gauranshmathur/arr_mcp/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/gauranshmathur/arr_mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/gauranshmathur/arr_mcp/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 2,625 This file is loaded in full into every session.

When invoked 2,625 The same file — it is already loaded in full.

Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.02625	$0.02625
Opus 5.5	$0.01050	$0.01050
Sonnet 5.5	$0.00525	$0.00525
Haiku 4.5	$0.00263	$0.00263

Measured 25d ago against content hash f1726af1e226, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

ARR_MCP 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 25d 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.

curl -s http://<host>:<port>/api/swagger.json | head -c 200

AGENTS.md · 139 lines

How it starts

The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Conventions for working in this repository. Read this before adding a service or a tool.

Non-negotiables

TDD. Write the test, run it, watch it fail for the right reason, then implement. Tests written after the code prove nothing — you never saw them catch anything.
Verify the API before implementing it. Never write a client from memory. See "Researching a service" below. Most of the bugs already fixed in this repo came from assuming an API shape.
Never write to stdout. Under the stdio transport, stdout carries the JSON-RPC stream. A single stray fmt.Println corrupts the framing and kills the session. Logging goes to stderr; TestLoggerDefaultsToStderr pins this.
Conventional commits. feat(scope):, fix(scope):, docs:, build:, ci:, test:, chore:. release-please builds the changelog from these, so the subject line is user-facing. Explain why in the body.
One feature per branch, one branch per worktree. See "Worktrees".
One release per feature. See "Releasing".

Researching a service

Do this in order, and stop at the first that gives an authoritative answer:

Live OpenAPI/Swagger document. Some services expose one unauthenticated:
```
curl -s http://<host>:<port>/api/swagger.json | head -c 200
```
Bazarr does. Sonarr 4.x and Radarr 6.x do not — they answer 404 even with a valid key, and unauthenticated they answer 401 for every path including nonexistent ones, which disguises the 404. For those, use the upstream repo's spec and verify each endpoint against the live instance.
Probe the live instance. Response schemas are often missing from the spec even when the paths are documented. When a key is needed, run the request inside the pod so the credential never leaves it:
```
kubectl exec <pod> -- sh -c 'K=$(...); curl -s -H "X-API-KEY: $K" http://127.0.0.1:<port>/api/<path>'
```
context7 for published library/API documentation.

Record anything surprising in the commit body. Real examples already found: Prowlarr serves /api/v1 while Sonarr and Radarr serve /api/v3; Bazarr wraps most responses in data but /badges and /system/languages are bare; provider resources (indexers, download clients, notifications) embed credentials in a fields array that must never reach a tool result.

Read the full file on GitHub · 139 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.

25d ago Changed · +54 tokens per session f1726af1e226
26d ago Changed · +2 lines · +203 tokens per session d45a55e9212e
1mo ago Changed · +67 tokens per session afa31927bd61
2mo ago First seen · 137 lines · 2,301 tokens per session scan A 7a5a395bf221

Subscribe to this mod's changes

ARR_MCP AGENTS.md is an instructions file published in the GitHub repository GauranshMathur/ARR_MCP (6 stars, last pushed 19d ago), licensed MIT. It adds 2,625 tokens to every session, about $0.0105 per session on Opus 5.5. 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.

arr media model-context-protocol self-hosted

ARR_MCP AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
ARR_MCP AGENTS.mdGauranshMathur/ARR_MCP	Instructions file	6	19d ago	MIT	A	2,625	Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

ARR_MCP: Instructions file for Codex