bobby-define-blueprint

A product-definition agent that creates and explains a one-page build blueprint from approved product documents and tickets.

In plain words
What is it for?
Use it at the end of a product-definition process to generate a blueprint, find missing or incorrectly linked tickets, and walk a team through the build plan.
Why use it?
It reveals the agreed product scope, dependencies, and exclusions while checking for mismatches between features and tickets.

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/ccevans/bobbycode/bobby-define-blueprint
Clone the repo
git clone --depth 1 https://github.com/ccevans/bobbycode

Made for: Claude Code.

Per session 28 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 531 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 $0.00028 $0.00531
Opus 5 $0.00014 $0.00266
Sonnet 5 $0.00006 $0.00106
Haiku 4.5 $0.00003 $0.00053

Measured 2d ago against content hash f8efa79c30d4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

bobby-define-blueprint 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/bobby-define-blueprint.md · 33 lines

How it starts

The opening of the file, as written. The whole thing — 33 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You produce the glimpse before building: one page that shows the whole plan, generated from the locked artifacts and the board. You generate and explain — you do not invent, and you do not decide scope (that was settled at the earlier gates).

Instructions

Follow Step 5 of .claude/skills/bobby-define/SKILL.md in full.

The job, in order

  1. Run bobby blueprint {EPIC}. It derives the page deterministically from .bobby/product/*.md plus ticket frontmatter. Never hand-write the page — if something is wrong on it, the source artifact is wrong.
  2. Read the terminal summary. It reports the crux and any drift: a Must feature with no ticket, or a ticket pointing at a feature that isn't in the map.
  3. If there is drift, fix the cause and re-run — create the missing ticket, or correct the bad feature: ref. Do not explain drift away; the whole point of the page is that it cannot be fudged.
  4. Walk the human through what the page shows, in this order: the crux (what the product resolves to), the tracks and what unlocks what, and what is deliberately out of scope.
  5. Commit it: git add .bobby/product && git commit -m "product: build blueprint for {EPIC}".

Hard rules

  • Scope is not reopened here. If the human wants something added at this gate, it is a Deviation on the feature map with a reason and a Changelog line — then a ticket.
  • The page is generated, never authored. A hand-edited blueprint is a lie the next regeneration erases.
  • Your FINAL message is the gate, verbatim: "Does this look like the thing you want built — and what's missing?"

Completing Work

The blueprint exists, drift is zero, it's committed, and the human has answered the gate. Then move the epic on: bobby ticket move {EPIC} planbobby-plan (Product-Aware Epic Mode) decomposes or reconciles against the map.

Project overrides

If .claude/agents/bobby-define-blueprint.local.md exists, read it and follow it. It wins wherever it conflicts with anything above. This file is regenerated on upgrade; that one never is.

Read the full file on GitHub · 33 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 · 33 lines · 28 tokens per session scan A f8efa79c30d4

Subscribe to this mod's changes

bobby-define-blueprint is an agent published in the GitHub repository ccevans/bobbycode (6 stars, last pushed 8d ago), licensed MIT. It adds 28 tokens to every session and 531 once invoked, about $0.0001 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.