Borrowing it
Nothing to install: this file belongs to richardthe3rd/cambridge-beer-festival-app. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/richardthe3rd/cambridge-beer-festival-app/main/.claude/skills/reference/SKILL.mdgit clone --depth 1 https://github.com/richardthe3rd/cambridge-beer-festival-appWrote 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/richardthe3rd/cambridge-beer-festival-app/reference)<a href="https://agentmods.dev/skills/richardthe3rd/cambridge-beer-festival-app/reference"><img src="https://agentmods.dev/badge/skills/richardthe3rd/cambridge-beer-festival-app/reference/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/richardthe3rd/cambridge-beer-festival-app/reference"><img src="https://agentmods.dev/badge/skills/richardthe3rd/cambridge-beer-festival-app/reference.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00211 | $0.07787 |
| Opus 5.5 | $0.00084 | $0.03115 |
| Sonnet 5.5 | $0.00042 | $0.01557 |
| Haiku 4.5 | $0.00021 | $0.00779 |
Grade A, and why
reference 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 2mo 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 — 461 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reference: Domain Theory for the Cambridge Beer Festival App
This is background knowledge, not a runbook. It explains what things mean so you can read code, triage a bug, or read a review comment without having to reverse-engineer intent from field names. For "how do I run/build/fix this" go to the sibling skills named throughout and in "When NOT to use" at the end.
1. The domain: what the festival actually is
The Cambridge Beer Festival is a real-ale festival organised in the
tradition of CAMRA (the Campaign for Real Ale) — the app's own festival
descriptions call it "the longest-running CAMRA beer festival in the UK
(since 1974)" (data/festivals.json, cbf2025.description) and the Play
Store listing points users to "the official Cambridge CAMRA website" for
authoritative festival info (docs/tooling/play-store.md:90). This app
itself is an unaffiliated community app — "not affiliated with or endorsed
by the Cambridge Beer Festival organizers" (docs/tooling/play-store.md:88)
— built and maintained independently of CAMRA.
Two editions exist, both modelled as Festival entries in the registry (see
§4):
- Main festival (
cbf{year}) — the big summer event, roughly a week long, on Jesus Green, Cambridge (data/festivals.json:cbf2026runs 2026-05-18..05-23). Full beverage range: beer, international beer, cider, perry, mead, wine, apple juice, low/no-alcohol. - Winter festival (
cbfw{year}) — a smaller winter edition at the Corn Exchange (cbfw2025: 2025-12-10..12-13,available_beverage_types: [beer, low-no]only — a genuinely narrower catalogue, not a data gap).
Three operational facts drive real design decisions in this repo — they are not incidental:
- Burst usage, one week a year. The app's entire yearly value is
concentrated into ~6 days. There is no ops capacity during the festival
(one maintainer, often on-site) — this is why the project enforces a
pre-festival deploy freeze (skill
change-control§5) and a free-tier-only infrastructure rule (same skill §4 Rule 2). - The venue has patchy, mobile-only data. Confirmed both by the
maintainer and in writing:
docs/planning/rating-service/design.md:18states the offline design choice ("Optimistic local + background sync") is because "Festival venues have patchy signal." Jesus Green and the Corn Exchange are marquees/halls with no reliable venue WiFi — attendees are on mobile data, often congested (thousands of people, one cell tower). This is the design-driving constraint behind the whole SWR (stale-while-revalidate, see §6 glossary) caching architecture: render cached data instantly, refresh in the background, never block the UI on a network round-trip, keep showing stale data with a dismissible notice rather than an error screen. See skillarchitecture-contractfor the mechanism and its historical bug family (#302,#306–#309). - Volunteers update availability ad hoc.
status_text(see §2) is typed in free text by festival volunteers at the bar as casks run low or arrive — not from a structured inventory system. That is why the vocabulary is inconsistent, why it can change between festivals (the_statusMapcarries anarrivedentry — historical winter vocabulary per issue #348/#349 records, not reproduced in the current live feeds), and why the parser cannot assume an ordinal/structured field exists (§3).
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.
- 2mo ago First seen · 461 lines · 211 tokens per session scan A 1dc70d0e0f2e
reference is a skill published in the GitHub repository richardthe3rd/cambridge-beer-festival-app (3 stars, last pushed today), licensed MIT. It adds 211 tokens to every session and 7,787 once invoked, about $0.0008 per session on Opus 5.5. 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.
reference compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| referencerichardthe3rd/cambridge-beer-festival-app | Skill | 3 | today | MIT | A | 211 | Claude Code |
| hr-onboardingnexu-io/open-design | Skill | 106k | yesterday | Apache-2.0 | A | 62 | Claude Code, Codex |
| book-mirrorgarrytan/gbrain | Skill | 32k | yesterday | MIT | A | 138 | Claude Code |
| exam-forecastanthropics/claude-for-legal | Skill | 9.9k | 3d ago | Apache-2.0 | A | 73 | Claude Code |
| lean-formalizewanshuiyin/Auto-claude-code-research-in-sleep | Skill | 19k | 3d ago | MIT | A | 81 | Claude Code |
| eli5companion-inc/feynman | Skill | 10k | 6d ago | MIT | A | 63 | Claude Code, Codex |
| deck-course-modulenexu-io/html-anything | Skill | 9.4k | 25d ago | Apache-2.0 | A | 25 | Claude Code, Codex |