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/alexei-led/cloud-mcp-server/claude-mdgit clone --depth 1 https://github.com/alexei-led/cloud-mcp-serverWrote 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/alexei-led/cloud-mcp-server/claude-md)<a href="https://agentmods.dev/instructions/alexei-led/cloud-mcp-server/claude-md"><img src="https://agentmods.dev/badge/instructions/alexei-led/cloud-mcp-server/claude-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 | $0.01114 | $0.01114 |
| Opus 5 | $0.00557 | $0.00557 |
| Sonnet 5 | $0.00223 | $0.00223 |
| Haiku 4.5 | $0.00111 | $0.00111 |
Grade C, and why
cloud-mcp-server CLAUDE.md scanned grade C 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 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.
Reaches for credential fileshighPrivilege escalation
SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.
- `aws://profiles` — parses `~/.aws/credentials` and `~/.aws/config` How it starts
The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AWS MCP Server Development Guide
Build & Test Commands
- Install all deps:
uv pip install --system -e ".[dev]" - Run tests:
make testorpython -m pytest -v -m "not integration" --timeout=60 - Run single test:
pytest tests/path/to/test_file.py::test_function_name -v - Run linter:
ruff check src/ tests/ - Format code:
ruff format src/ tests/ - Run server:
python -m aws_mcp_server - Run with streamable-http:
AWS_MCP_TRANSPORT=streamable-http python -m aws_mcp_server - Run with sandbox disabled:
AWS_MCP_SANDBOX=disabled python -m aws_mcp_server - Update lockfile:
uv pip compile --system pyproject.toml -o uv.lock - Versioning:
setuptools_scmfrom Git tags — tag asv1.x.yto release
Architecture
Server (server.py)
- Two tools:
aws_cli_help(readOnly) andaws_cli_pipeline(destructive, openWorld) FastMCPinstance withinstructions,icons,SERVER_DESCRIPTION- Tools use
ToolAnnotationsfrommcp.typesfor tool metadata - Tool functions return typed dataclasses (
CommandResult,CommandHelpResult) - All tool failures raise
ToolError— FastMCP setsisError: truein protocol response (SEP-1303)
Sandbox (sandbox.py)
- Landlock LSM for filesystem isolation (Linux only)
- Seccomp-bpf for syscall filtering
- AWS env vars from
SANDBOX_AWS_ENV_VARSare passed through to sandboxed processes AWS_MCP_SANDBOX=disabledto bypass for development;requiredto fail hard if unavailable
CLI Executor (cli_executor.py)
- All commands validated before execution (must start with
aws) - Pipe support: commands can include
| jq,| grep, etc. - Timeout handling with configurable default (300s)
- Returns
{"status": "success"|"error", "output": "..."}dict
Transport & Lifecycle (__main__.py)
- Transports:
stdio(default),streamable-http(recommended for web),sse(deprecated) - On
stdio: background thread monitors for client disconnect viaselect.pollor parent PID - On
sse/streamable-http: binds to0.0.0.0in Docker,127.0.0.1otherwise
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.
- 5d ago First seen · 77 lines · 1,114 tokens per session scan C 65115bcc52e7
cloud-mcp-server CLAUDE.md is an instructions file published in the GitHub repository alexei-led/cloud-mcp-server (186 stars, last pushed 6mo ago), licensed MIT. It adds 1,114 tokens to every session, about $0.0056 per session on Opus 5. A static security scan graded it C with 1 finding (reaches for credential files). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
insideout-power AGENTS.md
Instructions for luthersystems/insideout-power, covering insideout — riley infrastructure agent, what riley can do for you, how to invoke riley, via remote mcp (preferred — use this first) and via agent skills.
deadline-cloud AGENTS.md
AGENTS.md instructions for aws-deadline/deadline-cloud, covering agents.md - aws deadline cloud client, project overview, repository structure, key commands and development.
awsmux AGENTS.md
Instructions for 0hardik1/awsmux, covering agents.md, what awsmux is, build, test, verify, layout and the safety model — invariants you must not weaken.
finfocus CLAUDE.md
Claude Code instructions for rshade/finfocus, covering claude.md, critical instructions, project overview, build commands and single package/test commands.
finfocus GEMINI.md
Gemini CLI instructions for rshade/finfocus, covering finfocus development guidelines, active technologies, project structure, commands and build.
finfocus copilot-instructions.md
Copilot instructions for rshade/finfocus, covering finfocus development instructions, build, test, and lint commands, architecture, dual-mode binary and core components.