What is s3-mcp AGENTS.md?

A project instruction file for an MCP server that connects to Amazon S3, a service for storing files and other data online. It describes the code layout, setup commands, required AWS credentials, and testing rules.

It is for making changes to the s3-mcp project, which exposes S3-related tools through MCP. It also guides local development and testing with uv, Docker, pytest, and AWS environment variables.

It tells an agent how this Python server is structured and how to install dependencies, run tests, build it, and avoid misleading editor errors before setup is complete.

How JakePeralta7/s3-mcp configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to JakePeralta7/s3-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/JakePeralta7/s3-mcp/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/JakePeralta7/s3-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/jakeperalta7/s3-mcp/agents-md/github.svg)](https://agentmods.dev/instructions/jakeperalta7/s3-mcp/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/jakeperalta7/s3-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/jakeperalta7/s3-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/jakeperalta7/s3-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/jakeperalta7/s3-mcp/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 1,218 This file is loaded in full into every session.

When invoked 1,218 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.

Model	Per session	Once invoked
Fable 5.1	$0.01218	$0.01218
Opus 5.5	$0.00487	$0.00487
Sonnet 5.5	$0.00244	$0.00244
Haiku 4.5	$0.00122	$0.00122

Measured 1mo ago against content hash e2f6a9c90ceb, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

s3-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 1mo 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 · 83 lines

How it starts

The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Commands

Install deps: uv sync --group dev (creates .venv; Python >= 3.12)
Tests: uv run pytest -q — single test: uv run pytest tests/test_tools.py -k name
Lint: uv run ruff check .
Format: uv run ruff format . (check only: uv run ruff format --check .)
Type check: uv run pyright
Build image: docker build -t s3-mcp:dev .
Run server: uv run python -m s3_mcp (fails fast with RuntimeError unless AWS_ACCESS_KEY_ID/AWS_SECRET_ACCESS_KEY are set)
Read-only mode: set S3_READ_ONLY=true to disable write tools (put_object, delete_object, copy_object, create_bucket, delete_bucket, presign_put)
Committed pyrightconfig.json points Pyright at .venv: until uv sync has run, editors show false "Import could not be resolved" errors for boto3/mcp/pytest — install deps before trusting diagnostics.
Dev shell is Windows PowerShell: inline python -c "..." breaks on quoting — write a temp .py file instead.

Layout / architecture

The whole server is one module, src/s3_mcp/server.py. Keep it that way: env parsing (load_settings) -> boto3 client factory (create_client) -> plain do_* helpers -> 11 @mcp.tool registrations -> main().
Entrypoint chain: python -m s3_mcp -> __main__.py -> main() builds ONE boto3 client from env, then mcp.run(transport="stdio").
Tools are thin wrappers over do_* helpers that take the client as first arg. New tools must follow this split; tests call helpers directly with a MagicMock client and never go through async MCP machinery.
The client/region live in module globals (_client, _region), set by main() or injected in tests via server.set_client(client, region). Never build clients at import time.
The stdlib module is imported as import base64 as b64 because put_object has a parameter literally named base64.
Correlation IDs: Each tool call gets a 12-char hex correlation ID (uuid.uuid4().hex[:12]) for request tracing. Included in structured logs.
Read-only mode: Controlled by S3_READ_ONLY env var. When enabled, write tools are not registered at all (hidden from MCP clients).
Structured logging: Logs include correlation_id, tool name, and sanitized arguments (secrets redacted). Output goes to stderr for stdio transport.

Read the full file on GitHub · 83 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.

1mo ago Changed · +15 lines · +283 tokens per session e2f6a9c90ceb
2mo ago First seen · 68 lines · 935 tokens per session scan A 1fd43c5514ae

Subscribe to this mod's changes

s3-mcp AGENTS.md is an instructions file published in the GitHub repository JakePeralta7/s3-mcp (0 stars, last pushed 2d ago), licensed MIT. It adds 1,218 tokens to every session, about $0.0049 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 2026-08-31.

s3-mcp AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
s3-mcp AGENTS.mdJakePeralta7/s3-mcp	Instructions file	0	2d ago	MIT	A	1,218	Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	today	MIT	A	7,461	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	5,685	GitHub Copilot
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	3,093	GitHub Copilot
vscode model-prompts.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	2,022	GitHub Copilot

s3-mcp: Instructions file for Codex