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 skills/hannesill/m4/m4-apinpx skills add hannesill/m4 --skill m4-apigit clone --depth 1 https://github.com/hannesill/m4Wrote 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/skills/hannesill/m4/m4-api)<a href="https://agentmods.dev/skills/hannesill/m4/m4-api"><img src="https://agentmods.dev/badge/skills/hannesill/m4/m4-api.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 | $0.00038 | $0.01710 |
| Opus 5 | $0.00019 | $0.00855 |
| Sonnet 5 | $0.00008 | $0.00342 |
| Haiku 4.5 | $0.00004 | $0.00171 |
Grade A, and why
m4-api 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 — 168 lines — stays where its author put it; the contents beside it link to each section on GitHub.
M4 Python API
The M4 Python API provides programmatic access to clinical datasets for code execution environments. It mirrors the MCP tools but returns native Python types (DataFrames, dicts) instead of formatted strings.
When to Use the API vs MCP Tools
Use the Python API when:
- Complex clinical analysis - Multi-step analyses that require intermediate results, joins across queries, or statistical computations
- Large result sets - Query results with thousands of rows can be stored in DataFrames without dumping into context
- Mathematical operations - Aggregations, percentile calculations, statistical tests, and counting that benefit from pandas/numpy
- Iterative exploration - Building up analysis through multiple queries where each step informs the next
Use MCP tools when:
- Simple one-off queries where the result fits comfortably in context
- Interactive exploration where you want to see results immediately
Required Workflow
You must follow this sequence:
- Choose a dataset name and pass it explicitly, or create
M4Client(dataset=...) get_schema(dataset=...)/get_table_info(..., dataset=...)- Explore available tablesexecute_query()- Run SQL queries
from m4 import get_schema, get_table_info, execute_query
dataset = "mimic-iv" # or "mimic-iv-demo", "eicu", "mimic-iv-note"
# Step 1: Explore schema
schema = get_schema(dataset=dataset)
print(schema['tables']) # List of table names
# Step 2: Inspect specific tables before querying
info = get_table_info("mimiciv_hosp.patients", dataset=dataset)
print(info['schema']) # DataFrame with column names, types
print(info['sample']) # DataFrame with sample rows
# Step 3: Execute queries
df = execute_query(
"SELECT gender, COUNT(*) as n FROM mimiciv_hosp.patients GROUP BY gender",
dataset=dataset,
)
# Returns pd.DataFrame - use pandas operations freely
API Reference
Dataset Management
| Function | Returns | Description |
|---|---|---|
list_datasets() |
list[str] |
Available dataset names |
M4Client(dataset=...) |
M4Client |
Preferred explicit client for one dataset |
client.with_dataset(name) |
M4Client |
New client with the same session context and a different dataset |
client.switch_dataset(name) |
M4Client |
Mutate a client to another dataset for notebook-style sessions |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 168 lines · 38 tokens per session scan A a6ae7b2497b3
m4-api is a skill published in the GitHub repository hannesill/m4 (43 stars, last pushed 1mo ago), licensed MIT. It adds 38 tokens to every session and 1,710 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
experiment_management
Set up and manage the experiment folder structure. This is Phase 0 — it runs before any analysis begins. All bookkeeping files are JSON (never markdown).
evaluate
Compare baseline and new implementation results. Produce the machine-readable final report result.json, update experiments.json, and append a row to comparison.json.
implement
Create a new Jupyter notebook implementing the method from the research paper, using the same data as the baseline. Record implementation details and measured metrics as a structured JSON entry.
progress
Maintain a machine-readable progress file so dashboards, CLIs, and notebooks can poll the experiment's state at any time. The file is a JSON document — never markdown, never human-prose-first.
optimize
Drive structured autoresearch iteration after evo:discover and the baseline commit. Use when the user invokes /evo:optimize or asks to try ideas, try variants, run experiments, use available GPUs, improve the current best/frontier, continue an evo search, or compare candidate changes in an evo workspace. The…
arbor-agent-orchestrator
Top-level controller for recreating the open-source AutoResearch workflow as a suite of skills. Use when the user asks to run, emulate, extract, validate, or refine Arbor/AutoResearch behavior, especially when a coordinator must load phase skills for setup, ideation, executors, merge evaluation, novelty search…