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/itswendell/palot/agents-mdgit clone --depth 1 https://github.com/ItsWendell/palotWrote 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/itswendell/palot/agents-md)<a href="https://agentmods.dev/instructions/itswendell/palot/agents-md"><img src="https://agentmods.dev/badge/instructions/itswendell/palot/agents-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 | $0.02603 | $0.02603 |
| Opus 5 | $0.01301 | $0.01301 |
| Sonnet 5 | $0.00521 | $0.00521 |
| Haiku 4.5 | $0.00260 | $0.00260 |
Grade A, and why
palot AGENTS.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 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
The main process runs in Node.js, the renderer runs in a Chromium sandbox. They communicate via IPC only. Never import Node.js modules (`fs`, `child_process`, `path`) in the renderer -- use the `window.palot` bridge or ` How it starts
The opening of the file, as written. The whole thing — 194 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Palot Agent Instructions
Purpose of This File
This file is injected into every agent session for this project. Keep it short. Only add entries here if an agent is likely to get stuck or repeat a mistake without them. Do NOT add one-time setup notes, general knowledge, or things discoverable from config files.
Project Structure
- Monorepo: Turborepo + Bun workspaces (Bun 1.3.8)
packages/ui: Shared shadcn/ui component library (@palot/ui)packages/configconv: Universal agent config converter library (@palot/configconv) -- converts between Claude Code, OpenCode, and Cursor formatspackages/configconv-cli: Thin CLI wrapper (configconv) for the converter libraryapps/desktop: Electron 40 + Vite + React 19 desktop app (viaelectron-vite)apps/server: Bun + Hono backend -- used only in browser-mode dev (dev:web), NOT bundled with Electron
Desktop App Layout (apps/desktop/src/)
main/-- Electron main process (Node.js): window management, IPC handlers, OpenCode server lifecycle, filesystem readspreload/-- Electron preload bridge: exposeswindow.palotAPI viacontextBridgerenderer/-- React app (browser context): components, hooks, services, atoms (Jotai)
Skills
Project-specific skills live in .agents/skills/. Load a skill before starting
work that matches its domain -- they contain patterns and footguns that override
generic knowledge.
| Skill | When to load |
|---|---|
react-best-practices |
Writing or reviewing renderer components, optimizing re-renders or bundle size |
Commands
- Electron dev:
cd apps/desktop && bun run dev(electron-vite, renderer on port 1420) - Browser-only dev:
cd apps/desktop && bun run dev:web(Vite only, needsapps/serverrunning) - Backend server (browser mode only):
cd apps/server && bun run dev(port 3100) - Lint check:
bun run lint(from root) - Lint/format fix:
bun run lint:fixorbunx biome check --write .(from root) - Type check all:
bun run check-types(from root, via Turborepo) - Type check desktop:
cd apps/desktop && bun run check-types(usestsgo) - Run all tests:
cd packages/configconv && bun test - Run single test file:
cd packages/configconv && bun test test/converter/config.test.ts - Run tests by name:
cd packages/configconv && bun test --grep "converts model" - Rebuild server types:
cd apps/server && bun run build:types(required after adding server routes) - Add UI component:
cd packages/ui && bunx shadcn@latest add <component> - Package:
cd apps/desktop && bun run package(orpackage:linux,package:mac,package:win,package:all) - Package without code signing (macOS):
CSC_IDENTITY_AUTO_DISCOVERY=false cd apps/desktop && bun run package:mac - Changeset -- add:
bun changeset(interactive -- pick packages, bump type, write description) - Changeset -- version:
bun run version-packages(applies pending changesets, bumps versions, updates changelogs)
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 · 194 lines · 2,603 tokens per session scan A 334695bd5bd6
palot AGENTS.md is an instructions file published in the GitHub repository ItsWendell/palot (178 stars, last pushed 3mo ago), licensed MIT. It adds 2,603 tokens to every session, about $0.0130 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
prospec AGENTS.md
Instructions for benwu95/prospec, covering prospec, tech stack, language policy, core resources and constitution.
prospec CLAUDE.md
Instructions for benwu95/prospec, covering prospec, tech stack, language policy, core resources and constitution.
drenn CLAUDE.md
Instructions for anhnv02/drenn, covering claude.md, commands, architecture, configuration and src/main/ — electron main process.
blueprint CLAUDE.md
Claude Code instructions for Potato7711/blueprint: This project uses Blueprint, an engineering system for AI coding agents.
RemoClaw CLAUDE.md
Instructions for quangtam/RemoClaw, covering remoclaw — agent context, what is this?, architecture, file structure and how to run.
detach AGENTS.md
AGENTS.md instructions for iltsarev/detach, covering detach agent guide, start here, context and specification policy, context map and verification loop.