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/abrahao-dev/podium-app/claude-mdgit clone --depth 1 https://github.com/abrahao-dev/podium-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/instructions/abrahao-dev/podium-app/claude-md)<a href="https://agentmods.dev/instructions/abrahao-dev/podium-app/claude-md"><img src="https://agentmods.dev/badge/instructions/abrahao-dev/podium-app/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 | $0.01249 | $0.01249 |
| Opus 5 | $0.00624 | $0.00624 |
| Sonnet 5 | $0.00250 | $0.00250 |
| Haiku 4.5 | $0.00125 | $0.00125 |
Grade A, and why
podium-app 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 — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Podium
Infinite spatial canvas for terminal windows running AI coding agents (Claude Code, Codex, OpenCode, plain shells) plus markdown sticky notes. Terminals connect to terminals (agent-to-agent prompt piping via PTY stdin) and to notes (output capture). No built-in AI — orchestrates external CLI agents the user has installed.
Primary target: Windows. Development on macOS. Also supported: Linux (X11/Wayland via WebKitGTK). All three must build and run at all times.
Stack (fixed — do not substitute)
- Tauri 2 (Rust backend) + React 18 + TypeScript strict + Vite
- xterm.js (
@xterm/xterm+@xterm/addon-fit+@xterm/addon-webgl) portable-pty(Rust) — ConPTY on Windows, openpty on Unix- zustand (frontend state)
- React Flow (
@xyflow/react) for canvas/nodes/edges - Persistence: plain JSON in Tauri
app_data_dir(no database)
Keep deps minimal. Never update pinned versions once installed.
PTY IPC contract
Rust side: src-tauri/src/pty.rs — all platform-specific code lives in this module only.
Tauri commands (frontend wrapper: src/lib/pty.ts):
| Command | Args | Notes |
|---|---|---|
create_terminal |
id, cwd?, command?, cols, rows |
Spawns default shell; if command given, writes command + "\r" into stdin after spawn (PATH resolution via shell, robust cross-platform) |
write_to_terminal |
id, data |
Raw bytes to PTY stdin. Also powers agent-to-agent connections |
resize_terminal |
id, cols, rows |
|
kill_terminal |
id |
Events (Rust → frontend):
terminal-output:{id}— payloadVec<u8>(raw bytes; frontend feeds Uint8Array straight to xterm so its decoder handles split UTF-8)terminal-exit:{id}— payloadOption<u32>exit code. Session self-removes from the manager map on EOF.
One dedicated reader thread per session, 4KB reads.
Default shell: $SHELL on Unix; if unset, first existing of /bin/bash → /bin/zsh → /bin/sh (bash is ubiquitous on Linux, zsh the macOS default). Windows prefers pwsh.exe on PATH, else powershell.exe.
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 · 81 lines · 1,249 tokens per session scan A 4cc0d74c9cec
podium-app CLAUDE.md is an instructions file published in the GitHub repository abrahao-dev/podium-app (11 stars, last pushed 1mo ago), licensed MIT. It adds 1,249 tokens to every session, about $0.0062 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 2026-08-30.
Other instructions, from other repositories
plano CLAUDE.md
Claude Code instructions for zqkra/plano, covering claude.md, what plano is, commands, environment gotchas (windows — hit these during setup) and environment gotchas (linux / fedora kde — port-specific).
termcanvas AGENTS.md
Instructions for lout33/termcanvas, covering agents.md, product intent, architecture overview, main process — main.js and terminal identity model.
atelier-studio CLAUDE.md
Instructions for tofunori/atelier-studio, covering atelier studio — règles pour agents, système de design — contraignant, règle rust-first — contraignante (décision 2026-08-16) and contraintes techniques.
atelier-studio AGENTS.md
Instructions for tofunori/atelier-studio, covering ce qu'il faut savoir (2 min), protocole (copier-coller), 1. vérifications (obligatoires avant tout build), si gallery/ touché (serveur ou assets éditeurs) and 2. tuer tout (l'ordre importe peu, l'exhaustivité oui).
framework AGENTS.md
AGENTS.md instructions for ai-driven-dev/framework, covering agents.md, behavior, communication, action and memory management.
moonproxy-desktop AGENTS.md
AGENTS.md instructions for MoonProxyHQ/moonproxy-desktop, covering moonproxy, 功能, 技术栈, 目录结构 and 配置与数据存储.