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/block/berd/agents-mdgit clone --depth 1 https://github.com/block/berdWhat 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 | $0.01538 | $0.01538 |
| Opus 5 | $0.00769 | $0.00769 |
| Sonnet 5 | $0.00308 | $0.00308 |
| Haiku 4.5 | $0.00154 | $0.00154 |
Grade A, and why
berd 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 yesterday.
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 — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidelines for agents working on Berd.
Berd is a standalone Tauri 2 + React 19 desktop. ACP is the main interface we use for the actual agent loop - creating and running sessions, finding available models, and setting configuration. When available, we work over ACP methods, but the UI can handle operations that are not yet in ACP or are client specific.
Layout
src/— React UI/features/shared codesrc-tauri/— Tauri shell that starts or resolvesgoose servesdk/— vendored@aaif/goose-sdkpackage and generated ACP typesdistro/— bundled app defaults and packaged distribution assetsgoose-backend.lock.json— pinned upstream Goose backend used by dev and bundlesacp-tools.lock.json— release-controlledpackage.json+package-lock.jsonthe managed ACP bridges are installed from withnpm ciscripts/ensure-local-goose.sh— managed local Goose checkout for devscripts/prepare-goose-sidecar.sh— stages the pinned or explicit Goose binary for Tauri bundlingscripts/update-goose-backend-lock.sh— resolves and records a new Goose backend pinscripts/update-acp-tools-lock.mjs— resolves and records a new managed ACP bridge pinsrc/features/berdctl/— berdctl command registrysrc-tauri/plugins/berdctl/— berdctl brokersrc-tauri/crates/berdctl/— bundled berdctl CLIdistro/skills/berd-help/references/berdctl.md— berdctl guidance agents read from the bundledberd-helpskill
Architectural laws
LAWS/ defines required product and user experience behavior. Before planning,
implementing, or reviewing behavior changes, read LAWS/README.md and every
law file relevant to the affected behavior. Laws take correctness precedence
over the current code and tests; when they disagree, change the implementation
and tests or explicitly propose a product-approved law change.
Startup assets
Startup artifact media is resolved by the Tauri backend and returned as local
cache paths. Renderer code should use getArtifacts() or
selectProjectPreviewArtifacts() from src/shared/api/artifacts.ts, then pass
paths through convertFileSrc(..., "asset") before rendering media. Do not
vendor startup media, fetch catalogs in the renderer, or construct CDN
URLs in UI 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.
- yesterday First seen · 134 lines · 1,538 tokens per session scan A b7530938cc4f
berd AGENTS.md is an instructions file published in the GitHub repository block/berd (828 stars, last pushed 2d ago), licensed Apache-2.0. It adds 1,538 tokens to every session, about $0.0077 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
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.
buildNext
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).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
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).
spec-kit 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.
langchain 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.