trem-python

A Python-focused version of the TREM code-quality guide: Testable, Readable, Extensible, and Maintainable.

In plain words
What is it for?
It is for reviewing, refactoring, designing, or generating Python code using tools and patterns such as uv, FastAPI, Pydantic, SQLAlchemy, and FastMCP.
Why use it?
It gives Python projects consistent review and design guidance for code that should be isolated in tests, clear to read, safe to extend, and easier to change.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/clivechung/skillmcp/trem-python
Any agent
npx skills add clivechung/skillmcp --skill trem-python
Clone the repo
git clone --depth 1 https://github.com/clivechung/skillmcp

Made for: Claude Code, Codex.

Per session 88 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,362 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00088 $0.02362
Opus 5 $0.00044 $0.01181
Sonnet 5 $0.00018 $0.00472
Haiku 4.5 $0.00009 $0.00236

Measured yesterday against content hash 3c5c054bfb4e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

trem-python 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 yesterday.

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/skills/trem-python/SKILL.md · 163 lines

How it starts

The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TREM Python Engineering Framework

A tailored Python implementation of the TREM engineering framework (Testable, Readable, Extensible, Maintainable) optimized for the modern Python ecosystem.


🏗️ Standard Python Technology Stack

Capability Standard Tool / Library Key TREM Best Practice
Package Manager uv Declarative pyproject.toml, reproducible uv.lock, fast virtual environments with uv run and uv add.
Configuration pydantic / pydantic-settings Strongly-typed BaseSettings, environment variable loading, immutable configurations, validated input boundaries.
Logging Standard library logging Module-level logger = logging.getLogger(__name__), structured log formatters, lazy formatting (%s), never print().
CLI Applications typer Type-annotated CLI interfaces, dependency-injected helpers, tested with typer.testing.CliRunner.
REST APIs fastapi Dependency Injection via Depends(), Pydantic request/response models, isolated APIRouter modules, async I/O.
Database & Migrations alembic + sqlalchemy (Async) Declarative models, repository/unit-of-work patterns, automated schema migrations with Alembic, injected async sessions.
MCP Servers fastmcp (or official MCP SDK) Modular tool/resource definitions, decoupled business logic, strict type annotations, structured error returns.
Testing pytest + pytest-asyncio Fixtures with explicit scope, unittest.mock / AsyncMock, httpx.AsyncClient(transport=httpx.ASGITransport(app=app)) for warning-free ASGI testing.

⚡ Python TREM Assessment Checklist

Pillar Python-Specific Focus Verification Question
T (Testable) FastAPI Depends, Constructor Injection, Pytest Fixtures Are database sessions, HTTP clients, and clocks injected via arguments/Depends rather than instantiated globally or in function bodies?
R (Readable) Type Hints (PEP 484/585), Guard Clauses, Docstrings Are all function signatures strictly type-hinted? Is cyclomatic complexity minimized with early returns and descriptive variable names?
E (Extensible) Protocols (typing.Protocol), Pydantic Models, Strategies Are service interfaces defined with Protocol or ABCs allowing pluggable implementations (e.g. storage providers, notification backends)?
M (Maintainable) Layer Separation, Domain Exceptions, Alembic Migrations Are routers/CLIs separated from business logic and database access? Are custom exceptions handled uniformly without bare except:?

Read the full file on GitHub · 163 lines

Files

What ships with it

5 files 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.

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. yesterday First seen · 163 lines · 88 tokens per session scan A 3c5c054bfb4e

Subscribe to this mod's changes

trem-python is a skill published in the GitHub repository clivechung/skillmcp (0 stars, last pushed yesterday), licensed MIT. It adds 88 tokens to every session and 2,362 once invoked, about $0.0004 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.