Langpad AGENTS.md

A set of project instructions for developing an Obsidian plugin that connects to AI agents such as Claude Code, Codex, and Gemini CLI. It documents the plugin’s architecture, data flow, agent protocol, and permission handling.

In plain words
What is it for?
Use it when building or reviewing the plugin’s chat, session, agent, ACP communication, permission, terminal, and error-handling code.
Why use it?
It gives developers a shared reference for how the plugin’s components and agent communication are expected to work.

Instructions file for CodexOpenCode

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 instructions/langcliteam/langpad/agents-md
Clone the repo
git clone --depth 1 https://github.com/LangcliTeam/Langpad

Made for: Codex, OpenCode.

Per session 4,215 This file is loaded in full into every session.
When invoked 4,215 The same file — it is already loaded in full.
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 $0.04215 $0.04215
Opus 5 $0.02107 $0.02107
Sonnet 5 $0.00843 $0.00843
Haiku 4.5 $0.00421 $0.00421

Measured 2d ago against content hash ae77f7abf9c3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Langpad AGENTS.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 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.

AGENTS.md · 331 lines

How it starts

The opening of the file, as written. The whole thing — 331 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Client Plugin - LLM Developer Guide

Overview

Obsidian plugin for AI agent interaction (Claude Code, Codex, Gemini CLI, custom agents) via ACP.

Tech: React 19, TypeScript, Obsidian API, Agent Client Protocol (ACP)

Architecture

src/
├── types/                       # Type definitions (no logic, no dependencies)
│   ├── chat.ts                  # ChatMessage, MessageContent, PromptContent, AttachedFile, ActivePermission
│   ├── session.ts               # ChatSession, SessionUpdate (12-type union), SessionInfo, Capabilities
│   ├── agent.ts                 # AgentConfig, agent settings (Claude/Codex/Gemini/Custom)
│   └── errors.ts                # AcpError, ProcessError, ErrorInfo
├── acp/                         # ACP protocol (SDK dependency confined here)
│   ├── acp-client.ts            # Process lifecycle, UI-facing API (AcpClient class)
│   ├── acp-handler.ts           # SDK event handler + sessionId filter + listener broadcast
│   ├── type-converter.ts        # ACP SDK ↔ internal type conversion
│   ├── permission-handler.ts    # Permission queue, auto-approve, Promise resolution
│   └── terminal-handler.ts      # Terminal process create/output/kill
├── services/                    # Business logic (non-React, no React imports)
│   ├── vault-service.ts         # Vault access + fuzzy search + CM6 selection tracking
│   ├── settings-service.ts      # Reactive settings store (observer pattern only)
│   ├── session-storage.ts       # Session metadata + message file I/O (sessions/*.json)
│   ├── settings-normalizer.ts   # Settings validation helpers (str, bool, num, enumVal, etc.)
│   ├── session-helpers.ts       # Agent config building, API key injection (pure functions)
│   ├── session-state.ts         # Session state updates (legacy mode/model, config restore)
│   ├── message-state.ts         # Message array transforms (upsert, merge, streaming apply)
│   ├── message-sender.ts        # Prompt preparation + sending (pure functions)
│   ├── chat-exporter.ts         # Markdown export with frontmatter
│   ├── view-registry.ts         # Multi-view management, focus, broadcast
│   └── update-checker.ts        # Agent/plugin version checking
├── hooks/                       # React custom hooks (state + logic)
│   ├── useAgent.ts              # Facade: composes useAgentSession + useAgentMessages
│   ├── useAgentSession.ts       # Session lifecycle, config options, optimistic updates
│   ├── useAgentMessages.ts      # Message state, streaming (RAF batch), permissions
│   ├── useSuggestions.ts        # @[[note]] mentions + /command suggestions (unified)
│   ├── useSessionHistory.ts     # Session list/load/resume/fork
│   ├── useChatActions.ts        # Business callbacks (send, newChat, export, restart, etc.)
│   ├── useHistoryModal.ts       # Session history modal lifecycle
│   └── useSettings.ts           # Settings subscription (useSyncExternalStore)
├── ui/                          # React components
│   ├── ChatContext.ts           # React Context (plugin, acpClient, vaultService, settingsService)
│   ├── ChatPanel.tsx            # Orchestrator: calls hooks, workspace events, rendering
│   ├── ChatView.tsx             # Sidebar view (ItemView wrapper)
│   ├── FloatingChatView.tsx     # Floating window (position/drag/resize)
│   ├── ChatHeader.tsx           # Header (sidebar + floating variants)
│   ├── MessageList.tsx          # Virtualized message list (@tanstack/react-virtual)
│   ├── MessageBubble.tsx        # Single message rendering (content dispatch, copy button)
│   ├── ToolCallBlock.tsx        # Tool call + diff display (word-level highlighting)
│   ├── TerminalBlock.tsx        # Terminal output polling
│   ├── InputArea.tsx            # Textarea, attachments, mentions, history
│   ├── InputToolbar.tsx         # Config/mode/model selectors, usage, send button
│   ├── SuggestionPopup.tsx      # Mention/command dropdown
│   ├── PermissionBanner.tsx     # Permission request buttons
│   ├── ErrorBanner.tsx          # Error/notification overlay
│   ├── SessionHistoryModal.tsx  # Session history modal (list + confirm delete)
│   ├── FloatingButton.tsx       # Draggable launch button
│   ├── SettingsTab.ts           # Plugin settings UI
│   ├── view-host.ts             # IChatViewHost interface
│   └── shared/
│       ├── IconButton.tsx       # Icon button + Lucide icon wrapper
│       ├── MarkdownRenderer.tsx # Obsidian markdown rendering
│       └── AttachmentStrip.tsx  # Attachment preview strip
├── utils/                       # Shared utilities (pure functions)
│   ├── platform.ts              # Shell, WSL, Windows env, command building
│   ├── paths.ts                 # Path resolution, file:// URI
│   ├── error-utils.ts           # ACP error conversion
│   ├── mention-parser.ts        # @[[note]] detection/extraction
│   └── logger.ts                # Debug-mode logger
├── plugin.ts                    # Obsidian plugin lifecycle, settings persistence
└── main.ts                      # Entry point

Read the full file on GitHub · 331 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 · 331 lines · 4,215 tokens per session scan A ae77f7abf9c3

Subscribe to this mod's changes

Langpad AGENTS.md is an instructions file published in the GitHub repository LangcliTeam/Langpad (1 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 4,215 tokens to every session, about $0.0211 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.