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 instructions/clausd/aedificium-template/claude-mdgit clone --depth 1 https://github.com/clausd/aedificium-templateWrote 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/instructions/clausd/aedificium-template/claude-md)<a href="https://agentmods.dev/instructions/clausd/aedificium-template/claude-md"><img src="https://agentmods.dev/badge/instructions/clausd/aedificium-template/claude-md.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 | $0.03776 | $0.03776 |
| Opus 5 | $0.01888 | $0.01888 |
| Sonnet 5 | $0.00755 | $0.00755 |
| Haiku 4.5 | $0.00378 | $0.00378 |
Grade A, and why
aedificium-template CLAUDE.md 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 — 304 lines — stays where its author put it; the contents beside it link to each section on GitHub.
aedificium — Claude session guide
You are running as Claude Code inside a personal knowledge base for learning ideas — mathematics first of all, but any field. It is two things at once:
- a notebook of notes, discussions, worked exercises, and open questions, and
- a lexicon of definitions (a dictionary of terms and ideas).
A small Bun web app (server.ts) renders the texts on disk as a card grid and
chat-bridges to you via a Claude Code channel named aedificium. The user
reads the UI at http://localhost:8788, not your terminal output. To say
anything they will actually see, call the mcp__aedificium__reply tool — your
transcript does not reach them.
When the channel is policy-blocked (the case for some accounts), the chat-to-Claude direction does not deliver. The user then collaborates with you from the terminal, but the same conventions here still apply.
Math is core. Write LaTeX inline as $ … $ and display as $$ … $$. The
UI renders it with KaTeX. The markdown files are mirrored to GitHub for backup;
GitHub will not render the math, and that is fine — the files are the
database, the nice math view lives here.
Wiring
┌──────────────────────────────┐ channel notification
│ Browser UI (localhost:8788) │ ─────────────────────────────────▶
│ - card grid (left) │
│ - chat (right) │◀────────────────────────────────────
└──────────────────────────────┘ reply tool
▲
│ WebSocket
▼
server.ts (Bun)
│
▼
texts + pdfs on disk (source of truth)
notes/, lexicon/, pdfs/, assets/, files/, data/chat.jsonl
The source of truth is markdown files on disk. Everything is mirrored to GitHub (via the server's git auto-commit) purely as a backup; GitHub's web view is a fallback reader for the raw markdown.
Layout
notes/ free-form cards: discussions, exercises, questions,
YYYY-MM-DD-HHMM-slug.md scratchpad. Notes are allowed to be messy.
lexicon/ definitions — one term per file (a dictionary entry)
eigenvalue.md slug-named (no timestamp); frontmatter term/aka/domain
pdfs/ a library of PDFs for discussion + their sidecar cards
paper.pdf the binary (you do NOT embed a viewer)
paper.md auto-created sidecar: tags + running notes about paper.pdf
archive/ archived cards; preserves the original subdir
assets/ pasted/dropped images, referenced from cards
files/ misc non-PDF uploads from chat
data/
chat.jsonl durable chat transcript (tracked; mineable + searchable)
server.ts channel + UI
.mcp.json spawns server.ts for Claude Code
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 · 304 lines · 3,776 tokens per session scan A 45ef71fbfa17
aedificium-template CLAUDE.md is an instructions file published in the GitHub repository clausd/aedificium-template (0 stars, last pushed 15d ago), licensed MIT. It adds 3,776 tokens to every session, about $0.0189 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.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.