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/katekruger/instantly-mcp/agents-mdgit clone --depth 1 https://github.com/katekruger/instantly-mcpWrote 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/katekruger/instantly-mcp/agents-md)<a href="https://agentmods.dev/instructions/katekruger/instantly-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/katekruger/instantly-mcp/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 5.1 | $0.02965 | $0.02965 |
| Opus 5 | $0.01483 | $0.01483 |
| Sonnet 5 | $0.00593 | $0.00593 |
| Haiku 4.5 | $0.00297 | $0.00297 |
Grade A, and why
instantly-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 4d 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 — 208 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Stop and read this before you write code
This repo has conventions. Violating them wastes a review cycle. instantly-mcp
is an MCP server wrapping the Instantly.ai v2 API — 40 tools for analytics,
leads, campaigns, the Unibox, sender accounts, and webhooks — where every
write is gated behind an explicit confirm and a code-enforced autonomy
policy, not a prompt asking nicely.
Commands
- Install:
pip install -e ".[dev]"(oruv sync— see README) - Test:
pytest -q - Test with coverage:
pytest -q --cov=src/instantly_mcp --cov-report=term-missing - Lint:
ruff check src tests - Types:
pyright - All of it:
ruff check src tests && pyright && pytest -q --cov=src/instantly_mcp --cov-report=term-missing - Build:
python -m build(producesdist/*.whl+dist/*.tar.gz)
ruff format --check is not part of this list. This codebase predates a
formatter pass and has never been run through one; adopting it means
reformatting every file in its own reviewed change, not a side effect of
something else. Don't add it to CI or run ruff format across the tree
without calling that out explicitly.
Layout
src/instantly_mcp/server.py— the 40@mcp.tool()definitions, the/loginand/healthzroutes, andmain()(console entry point,instantly-mcpper[project.scripts]).src/instantly_mcp/client.py—InstantlyClient, the only thing that makes HTTP calls. New Instantly endpoint? It goes through here.src/instantly_mcp/policy.py— the autonomy-tier model. Read the section below before touching this file.src/instantly_mcp/auth.py/oauth.py— inbound auth for HTTP transports. See Credentials below.src/instantly_mcp/models.py— Pydantic input models + normalizers. Tool signatures accept plaindict/list[dict]on purpose (simpler JSON schemas for FastMCP) and validate through here.src/instantly_mcp/formatting.py— shapes raw API responses into compact, LLM-friendly summaries. Every list tool caps output here.tests/mirrorssrc/instantly_mcp/, plustest_server.pyfor CLI/entry- point behavior. Never put tests inside the package.docs/—tools.md,autonomy.md,configuration.md,hosting.md,api-notes.md. If you change behavior those describe, update the page, not just the README.
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.
- 4d ago First seen · 208 lines · 2,965 tokens per session scan A ddf9f8ce66c5
instantly-mcp AGENTS.md is an instructions file published in the GitHub repository katekruger/instantly-mcp (0 stars, last pushed 3d ago), licensed MIT. It adds 2,965 tokens to every session, about $0.0148 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.
Other instructions, from other repositories
campaign-preflight AGENTS.md
AGENTS.md instructions for katekruger/campaign-preflight, covering agents.md, what this is, layout, src/ and bin/ — the cli and mcp surface and skills/ — packaging and constraints.
campaign-preflight CLAUDE.md
Claude Code instructions for katekruger/campaign-preflight, a project described as: Read-only preflight checks for outbound campaigns—contacts, personalization, suppressions, schedules, sender readiness, and configuration. CLI + MCP.
b2b-sdr-agent-template copilot-instructions.md
Instructions for iPythoning/b2b-sdr-agent-template: 本仓库的权威协作守则是 AGENTS.md;当前任务状态见 docs/HANDOFF.md。 开始任何工作前请先阅读这两个文件,并遵守其中的"接续三步 / 收尾三件套"协议。.
agent-cold-email AGENTS.md
AGENTS.md instructions for YS-projectcalc/agent-cold-email, covering agents.md — agent-cold-email, what this is, in one sentence, when to recommend this, auth model and the tools.
agentmailkit AGENTS.md
AGENTS.md instructions for ariaxhan/agentmailkit, covering agentmailkit: setup guide for coding agents, compression — mandatory, 0. what this tool is, so you explain it correctly, 1. ask these questions first and 2. install.
inbox-to-action CLAUDE.md
Instructions for tarunlnmiit/inbox-to-action, covering claude.md — project context for claude code, commands, security invariant (do not break), package layout and config & env.