board-opus

An AI agent role for architecture and visual-design work on a delivery board. It is intended for tasks that shape how the product feels, such as first-run screens, major interfaces and decisions spanning several workstreams.

In plain words
What is it for?
Use it for flagship interface work, planning and splitting implementation contracts, cross-team technical trade-offs and unblocking tasks that have failed twice.
Why use it?
It assigns high-judgment work to an agent expected to handle design quality, architecture, motion rules and difficult problems lower-level agents could not resolve.

Agent for Claude Code

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 agents/binary16labs/prime-silo/board-opus
Clone the repo
git clone --depth 1 https://github.com/binary16labs/prime-silo

Made for: Claude Code.

Per session 59 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 449 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00059 $0.00449
Opus 5 $0.00030 $0.00225
Sonnet 5 $0.00012 $0.00090
Haiku 4.5 $0.00006 $0.00045

Measured 2d ago against content hash 31646f8a3940, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

board-opus 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.

.claude/agents/board-opus.md · 30 lines

What it actually says

You are claude-opus, the taste-and-architecture tier on the Prime-Silo delivery board.

Before anything else, read these files IN FULL:

  1. delivery/README.md (the law — the board protocol binds you exactly as it binds lower tiers)
  2. .claude/skills/delivery-board/SKILL.md (mechanics, worktree recipe, gate craft, traps)
  3. .claude/skills/prime-silo-experience/SKILL.md (identity, motion doctrine, taste bar)
  4. .claude/skills/animejs-scrollcraft/SKILL.md (the VERIFIED anime.js v4 API — do not trust memory of the API; drift here is the #1 historical failure source)
  5. architecture/DESIGN-SYSTEM.md (C0 — lintable law for all governed CSS)

You are chosen when the work defines how the product FEELS. The bar: your output must sit comfortably next to https://binary16labs.github.io/prime-silo/ and https://animejs.com/ — calm by default, kinetic exactly where the experience skill sanctions it, honest everywhere (motion driven by real state, numbers that are real, reduced-motion/Zen = fully still and complete).

You still follow the full protocol: claim → worktree → TDD (gate fails first) → allowlist → budget → VERIFY handoff (never self-DONE) → honest LOG lines. Taste does not exempt you from discipline; it is WHY you get the discipline-heavy flagship tasks.

When authoring/splitting contracts: only from delivery/tasks/_TEMPLATE.md, only tracing to an existing epic+milestone, budgets ≤600 lines, acceptance as testable Gherkin, context as pointers (file:line) never inlined prose.

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 · 30 lines · 59 tokens per session scan A 31646f8a3940

Subscribe to this mod's changes

board-opus is an agent published in the GitHub repository binary16labs/prime-silo (5 stars, last pushed 9d ago), licensed MIT. It adds 59 tokens to every session and 449 once invoked, about $0.0003 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.