termdat-mcp CLAUDE.md

termdat-mcp CLAUDE.md is an instructions file for coding agents from malkreide/termdat-mcp. It costs 6,303 tokens per session, scanned A, a copy of zurich-opendata-mcp CLAUDE.md, MIT.

A set of project instructions for Claude Code, including rules for keeping a code checkout current, running tests, and responding when checks fail. It also describes conventions for a portfolio of repositories.

In plain words
What is it for?
Use it when maintaining the listed repositories, especially before making changes, running tests, or investigating a failed build or code-quality check.
Why use it?
It helps the coding agent avoid working from an outdated copy of a repository or using checks that differ from the project's continuous-integration setup.

Instructions file

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 instructions/malkreide/termdat-mcp/claude-md
Clone the repo
git clone --depth 1 https://github.com/malkreide/termdat-mcp

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.

agentmods badge for termdat-mcp CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/malkreide/termdat-mcp/claude-md.svg)](https://agentmods.dev/instructions/malkreide/termdat-mcp/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/malkreide/termdat-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/malkreide/termdat-mcp/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 6,303 This file is loaded in full into every session.
When invoked 6,303 The same file — it is already loaded in full.
Security scan A 1 finding. Scan, not verified.
Origin 80% copy Near-identical to another mod 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 $0.06303 $0.06303
Opus 5 $0.03152 $0.03152
Sonnet 5 $0.01261 $0.01261
Haiku 4.5 $0.00630 $0.00630

Measured 5d ago against content hash 4aa8df5e9087, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

termdat-mcp CLAUDE.md scanned grade A 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

Token aus der Umgebung in einen curl-Header zu setzen, blockiert der
Origin

This is a copy

80% identical to zurich-opendata-mcp CLAUDE.md — 183 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

CLAUDE.md · 465 lines

How it starts

The opening of the file, as written. The whole thing — 465 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

Teil 1 — Portfolio-Konventionen

Vor der Arbeit

Klon-Aktualität prüfen — Standard-Branch ermitteln, nicht main annehmen:

B=$(git ls-remote --symref origin HEAD | sed -n 's|^ref: refs/heads/\([^[:space:]]*\).*|\1|p')
git fetch origin "${B:?Standard-Branch nicht ermittelbar}" &&
  git rev-list --count HEAD..FETCH_HEAD

Drei Server im Portfolio heissen ihren Standard-Branch master (openlex-mcp, swiss-courts-mcp, swisstopo-mcp); dort scheitert ein fest verdrahtetes origin/main mit «couldn't find remote ref main». Wer das für ein Netzproblem hält, arbeitet weiter auf genau dem veralteten Klon, vor dem dieser Absatz warnt. Den :?-Schutz nicht weglassen: Bei leerem B fetcht git still den Remote-HEAD und endet mit 0.

Ein veralteter Klon erzeugt eine rote CI, deren Ursache nicht im Diff steht. Am 3.8.2026 zweimal passiert — beide Male fehlten genau die Commits, die das Gate einführten, an dem der Branch scheiterte.

Gates lokal fahren, mit der GEPINNTEN ruff-Version aus der CI. Eine andere Version meldet Abweichungen, die niemand verursacht hat.

Tests

Gegenprobe ist Pflicht. Ein Test, der grün bleibt, wenn man die Implementierung entfernt, prüft nichts. Jede neue Zusicherung einzeln neutralisieren und zeigen, dass genau die zugehörigen Tests fallen.

Zwei Fallen, die beide grün blieben:

  • Eine Fake-Uhr, die nur beim Schlafen vorrückt, kann eine Zusicherung über echte Zeit nicht widerlegen.
  • monkeypatch.setattr(modul.asyncio, "sleep", ...) greift ins Modul asyncio selbst und entschärft die Mechanik im ganzen Prozess. Patche einen Modul-Alias (_sleep = asyncio.sleep), nicht das fremde Modul.

Handgeschriebene Fixtures kodieren die Annahme des Autors und können sie nicht widerlegen. Mindestens eine aufgezeichnete Antwort pro externem Endpunkt, mit Aufnahmedatum.

Wenn etwas rot ist

Roter Live-Test: erst die Quelle abfragen, dann einordnen. Nicht aus der Fehlermeldung schliessen. Am 3.8.2026 hiess "nicht gefunden" nicht, dass der Datensatz weg war, sondern dass die Quelle die Schreibweise ihrer Kopfzeile gewechselt hatte — vier von sechs Datensätzen produktiv kaputt, alle Unit-Tests grün.

Read the full file on GitHub · 465 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.

  1. 5d ago First seen · 465 lines · 6,303 tokens per session scan A 4aa8df5e9087

Subscribe to this mod's changes

termdat-mcp CLAUDE.md is an instructions file published in the GitHub repository malkreide/termdat-mcp (1 stars, last pushed 3d ago), licensed MIT. It adds 6,303 tokens to every session, about $0.0315 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). It is 80% identical to zurich-opendata-mcp CLAUDE.md, differing in 183 lines, and is treated as a copy.