cantrips-loop

A guide to six storage actions and the knowledge stores enabled in this repository. It uses temporary Markdown files for feature specifications and tickets, with finished results kept in code, git history and other knowledge stores.

In plain words
What is it for?
Use it when publishing, fetching or commenting on a specification, creating tickets, or following this repository's documented storage workflow.
Why use it?
It clarifies where working plans belong, how to read or update them, and what should remain after a feature is finished.

Agent

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 agents/citiesskylinesmodding/agents-plugins/cantrips-loop
Clone the repo
git clone --depth 1 https://github.com/CitiesSkylinesModding/agents-plugins
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 540 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 $0.00000 $0.00540
Opus 5 $0.00000 $0.00270
Sonnet 5 $0.00000 $0.00108
Haiku 4.5 $0.00000 $0.00054

Measured 2d ago against content hash a3d32bb2df0e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cantrips-loop 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 2d 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.

docs/agents/cantrips-loop.md · 26 lines

How it starts

The opening of the file, as written. The whole thing — 26 lines — stays where its author put it; the contents beside it link to each section on GitHub.

The cantrips loop in this repo

What the six storage verbs translate to here, and which knowledge stores are enabled. Storage-touching skills read this doc instead of the plugin defaults. /setup-cantrips-loop wrote it; re-run that skill to change it.

Storage backend

Local markdown under .scratch/<feature>/, a kebab-case feature slug per folder. .scratch/ is already listed in .gitignore, so a write there needs no setup.

A spec and its tickets are working material, not a record: they live in .scratch/ only while the feature is in flight, and the durable outcome lands in code, git history, and the knowledge stores below. Closing a finished feature is the human's act — they delete .scratch/<feature>/ once it no longer serves.

  • Publish the spec (publish-spec) — write the spec to .scratch/<feature>/spec.md.
  • Fetch the spec (fetch-spec) — read .scratch/<feature>/spec.md.
  • Annotate the spec (annotate-spec) — append the note under a ## Comments heading at the end of the spec file, each entry prefixed with its date; the body above that heading stays frozen.
  • Publish the tickets (publish-tickets) — write one file per ticket to .scratch/<feature>/NN-<slug>.md, numbered from 01 in dependency order, blocking edges as the ticket body's "Blocked by" prose; the shared folder is what ties a ticket to its parent spec. A ticket's body takes the shape docs/authoring/ticket-template.md fixes.
  • Fetch the ticket (fetch-ticket) — read the ticket file.
  • Resolve the ticket (resolve-ticket) — add or flip a Status: resolved line directly under the ticket's title; the verified acceptance-criterion checkboxes remain as evidence.

Three invariants hold whatever the backend: an annotation appends without touching what is already there and stays time-ordered; an annotation recording a revised decision governs over the body it revises, so later readers judge the spec as amended; and a published ticket is traceable to its parent spec.

Read the full file on GitHub · 26 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. 2d ago First seen · 26 lines · 0 tokens per session scan A a3d32bb2df0e

Subscribe to this mod's changes

cantrips-loop is an agent published in the GitHub repository CitiesSkylinesModding/agents-plugins (5 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 540 tokens. 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.