now-sdk-ext-mcp: Instructions file for Codex

AGENTS.md

now-sdk-ext-mcp AGENTS.md is an instructions file for Codex, OpenCode from sonisoft-cnanda/now-sdk-ext-mcp. It costs 1,151 tokens per session, scanned A, original, MIT.

Repository guidance for an MCP server that exposes ServiceNow tools to AI agents through JSON-RPC over standard input and output.

In plain words
What is it for?
Use it when adding tools, changing agent-facing descriptions, or debugging communication and credential-related behavior.
Why use it?
It highlights rules that prevent communication from breaking, such as keeping diagnostic messages off the protocol's output stream.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: the file is AGENTS.md. Also seen: mentions CLAUDE.md; mentions AGENTS.md.

This is sonisoft-cnanda/now-sdk-ext-mcp's own configuration. It tells Codex and OpenCode how to work on now-sdk-ext-mcp itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything now-sdk-ext-mcp configures →

Reuse

Borrowing it

Nothing to install: this file belongs to sonisoft-cnanda/now-sdk-ext-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/sonisoft-cnanda/now-sdk-ext-mcp/main/AGENTS.md
Clone the repo
git clone --depth 1 https://github.com/sonisoft-cnanda/now-sdk-ext-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.

agentmods badge for now-sdk-ext-mcp AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/sonisoft-cnanda/now-sdk-ext-mcp/agents-md/github.svg)](https://agentmods.dev/instructions/sonisoft-cnanda/now-sdk-ext-mcp/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/sonisoft-cnanda/now-sdk-ext-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/sonisoft-cnanda/now-sdk-ext-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.

agentmods 80×15 button for now-sdk-ext-mcp AGENTS.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/sonisoft-cnanda/now-sdk-ext-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/sonisoft-cnanda/now-sdk-ext-mcp/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,151 This file is loaded in full into every session.
When invoked 1,151 The same file — it is already loaded in full.
Security scan A 0 findings. 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.01151 $0.01151
Opus 5.5 $0.00460 $0.00460
Sonnet 5.5 $0.00230 $0.00230
Haiku 4.5 $0.00115 $0.00115

Measured 2mo ago against content hash e94c82d28cf6, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

now-sdk-ext-mcp 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 2mo 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.

AGENTS.md · 122 lines

How it starts

The opening of the file, as written. The whole thing — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Operating rules for automated agents working in this repository.

For architecture, layout and build commands, read CLAUDE.md. This file is only about the things that will bite you here and not elsewhere.


What this repository is

An MCP server exposing ~86 ServiceNow tools to AI agents. It is launched as a non-interactive child process, speaks JSON-RPC over stdio, and acts against real instances with the launching user's credentials.

Two things follow, and both are unusual enough to be worth stating:

  • stdout is a protocol channel, not an output channel.
  • The caller is a model, not a person. Tool descriptions and result text are the entire interface. Prose here is not documentation about the product; it is the product.

Hard rules

1. Never write to stdout

console.log, a stray print, a library that logs on import — any byte on stdout that is not JSON-RPC corrupts the stream and breaks the client's parser, usually with an error that points nowhere near the cause.

Diagnostics go to stderr, always:

console.error("[thing] message");

This is why credstore-boot.ts uses process.stderr.write and why the progress helper swallows notification failures to stderr rather than surfacing them.

2. A tool's description is its implementation, as far as the model is concerned

Tools are hand-registered with a title, a description, and a zod inputSchema. The model never reads the handler. If the description does not say that a flow with approval steps must run in background mode, the model will not know.

Two patterns already in the codebase, both worth continuing:

  • Explain the platform, not just the parameter. execute_script's description covers scoped-vs-global name qualification with concrete examples, because that is the thing callers actually get wrong.
  • Put next steps in the RESULT, especially on empty and error paths. "No tables found in this search group. Use \add_code_search_table` to add tables."` costs nothing until the model is stuck, and is how chains self-correct without bloating every description.

Read the full file on GitHub · 122 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. 2mo ago First seen · 122 lines · 1,151 tokens per session scan A e94c82d28cf6

Subscribe to this mod's changes

now-sdk-ext-mcp AGENTS.md is an instructions file published in the GitHub repository sonisoft-cnanda/now-sdk-ext-mcp (16 stars, last pushed 1mo ago), licensed MIT. It adds 1,151 tokens to every session, about $0.0046 per session on Opus 5.5. 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.

Related

now-sdk-ext-mcp AGENTS.md compared with instructions from other repositories

now-sdk-ext-mcp AGENTS.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
now-sdk-ext-mcp AGENTS.mdsonisoft-cnanda/now-sdk-ext-mcpInstructions file161mo agoMITA1,151Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143kyesterdayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149kyesterdayMITA4,680Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA5,516GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA3,093GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file274k2d agoMITA4,313Codex, OpenCode