Borrowing it
Nothing to install: this file belongs to DanWahlin/ai-agent-board. 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/DanWahlin/ai-agent-board/main/AGENTS.mdgit clone --depth 1 https://github.com/DanWahlin/ai-agent-boardWrote 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/danwahlin/ai-agent-board/agents-md)<a href="https://agentmods.dev/instructions/danwahlin/ai-agent-board/agents-md"><img src="https://agentmods.dev/badge/instructions/danwahlin/ai-agent-board/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 5.1 | $0.02948 | $0.02948 |
| Opus 5 | $0.01474 | $0.01474 |
| Sonnet 5 | $0.00590 | $0.00590 |
| Haiku 4.5 | $0.00295 | $0.00295 |
Grade A, and why
ai-agent-board 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 6d 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 — 160 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Project Overview
Agentic AI Kanban Board — a drag-and-drop Kanban board that delegates coding tasks to AI coding agents (GitHub Copilot, Claude Code, OpenAI Codex, OpenCode, Hermes, OpenClaw). Monorepo with npm workspaces.
Architecture
ai-agent-board/
├── packages/
│ ├── client/ # React 19 + Vite + Tailwind 4 + Framer Motion + xterm.js
│ │ └── src/
│ │ ├── components/ # Board, Column, TaskCard, TaskGroupCard, GroupPanel, AgentPanel, TerminalView, FilterChips, Header, dialogs
│ │ ├── hooks/ # useTasks, useTaskGroups, useTheme, useDebounce, useKeyboardShortcuts
│ │ ├── lib/ # api.ts (REST + WebSocket), agent-config.ts, priority-config.ts, columns.ts, utils
│ │ └── types/ # Client-side type re-exports
│ ├── server/ # Express + @codewithdan/agent-sdk-core + better-sqlite3/pg + ws
│ │ └── src/
│ │ ├── middleware/ # auth.ts (Bearer token auth)
│ │ ├── routes/ # tasks.ts, agent.ts, git.ts, templates.ts, groups.ts, helpers.ts
│ │ ├── services/ # agent-manager.ts (session orchestration, event caching)
│ │ ├── repositories/# sqlite.ts, postgres.ts, sqlite-templates.ts, postgres-templates.ts, sqlite-groups.ts, postgres-groups.ts, types.ts, template-types.ts, group-types.ts
│ │ ├── db.ts # SQLite + PostgreSQL init + migrations
│ │ ├── websocket.ts # WebSocket broadcast
│ │ └── index.ts # Express app setup + graceful shutdown
│ └── e2e/ # Playwright tests
├── shared/ # Shared types (Task, TaskGroup, TaskTemplate, AgentEvent, ColumnId, AgentType, etc.) + validation constants
├── scripts/ # required gate, hook install, and E2E process helpers
└── k8s/ # Kubernetes manifests (namespace, deployments, services, ingress)
Key Technical Decisions
- Multi-agent support — pluggable
AgentProvider/AgentSessioninterfaces from@codewithdan/agent-sdk-core. Six providers:copilot(@github/copilot-sdk),claude(@anthropic-ai/claude-agent-sdk),codex(@openai/codex-sdk),opencode(@opencode-ai/sdk),hermes, andopenclaw. Auto-detected at startup viadetectAgents(). - Agent SDK abstraction — all provider implementations live in the external
@codewithdan/agent-sdk-corepackage. The server imports providers and the detection function from this package. - Dual database backends — SQLite via
better-sqlite3(default, zero config) or PostgreSQL viapg(setDATABASE_URL). Both implement theTaskRepositoryandTemplateRepositoryinterfaces. - Route splitting — REST API is split across
tasks.ts(CRUD),agent.ts(start/stop/events/follow-up),git.ts(merge-local, create-pr, worktree cleanup, git-info),templates.ts(task template CRUD), andgroups.ts(group CRUD + run/stop/archive). - API key auth — optional Bearer token via
API_KEYenv var. When set, all API and WebSocket requests requireAuthorization: Bearer <key>. Middleware inmiddleware/auth.ts. - Task Groups — parent entity with N child tasks, concurrency-controlled execution via
GroupQueuein agent-manager. Groups move as a single card on the board; auto-advance to review when all children complete. Parallelism slider locked once running. - Event streaming — SDK events mapped to
AgentEvents, persisted to database, broadcast via WebSocket. In-memory LRU cache (200 tasks max, 100 events per task). - Git worktrees — optional per-task branch isolation. Agent works in worktree directory, path rewriting via
onPreToolUsehook. Worktrees auto-cleaned after successful merge or PR creation. - Local merge —
mergeLocal()merges worktree branch into base branch locally with per-repo mutex to prevent concurrent checkout races. Auto-aborts on conflict. - Smart PR/merge buttons —
GET /api/tasks/:id/git-infochecks for remote; UI shows "Create PR" only when remote exists, "Merge to main" always available. - Vite proxy — client proxies
/apiand/wsto the server. - Shared validation —
shared/constants.tsexports validators (isValidPriority,isValidColumnId, etc.) and limits (MAX_TITLE_LENGTH,MAX_DESCRIPTION_LENGTH) used by both client and server.
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.
- 6d ago First seen · 160 lines · 2,948 tokens per session scan A 54df743d77fc
ai-agent-board AGENTS.md is an instructions file published in the GitHub repository DanWahlin/ai-agent-board (57 stars, last pushed 11d ago), licensed MIT. It adds 2,948 tokens to every session, about $0.0147 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
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
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 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).
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.
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.