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 skills/owainlewis/blueprint/designnpx skills add owainlewis/blueprint --skill designgit clone --depth 1 https://github.com/owainlewis/blueprintWrote 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/skills/owainlewis/blueprint/design)<a href="https://agentmods.dev/skills/owainlewis/blueprint/design"><img src="https://agentmods.dev/badge/skills/owainlewis/blueprint/design.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 | $0.00055 | $0.01465 |
| Opus 5 | $0.00028 | $0.00732 |
| Sonnet 5 | $0.00011 | $0.00293 |
| Haiku 4.5 | $0.00006 | $0.00146 |
Grade A, and why
design 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 — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Design
Settle choices that implementation must not invent. A useful design states the behavior, the main tradeoff, and the proof before code is written.
Process
- Read the request, repository instructions, relevant code, and linked material.
- Identify choices that would change behavior, interfaces, data, errors, security, operations, or tests.
- Ask blocking questions before drafting. Ask only when the answer would change the design, and recommend an answer. Record non-blocking questions and a recommended default under Open questions.
- Write
docs/<feature-slug>/design.mdusing the numbered shape below. Keep it short and in order. Omit only sections that do not apply. - Run the review pass. Fix what you can. List the rest under Open questions.
- Stop with the proposed design ready for review. Do not plan or implement it.
Document shape
# <Title>
> **Status:** Proposed for review
## 1. Executive summary
Say what is wrong today, who feels the problem, what will change, how we plan to fix it, and the main downside. Use simple words. Do not list sections or implementation details.
## 2. Context and scope
Describe the current behavior, why it is insufficient, what changes once this ships, and the boundary of this design.
## 3. System context
Show where the change fits in the current system. Name the parts and outside systems it touches and the boundaries it must preserve. Include a small diagram when it makes those relationships clearer.
## 4. Proposed design
### How it works
Walk one real case from start to finish. Name the thing that arrives, what handles it, what gets written down, and what the user sees.
### Components and responsibilities
For each changed part, state what it owns, what it depends on, and what it does not own.
### Decisions
For each real choice, say what you chose, what you rejected, and what the choice costs. Use one short paragraph. Skip choices nobody would question.
## 5. Invariants and requirements
### Invariants
List rules that must always hold as `INV-1`, `INV-2`, and so on. A reviewer checks the code against these rules, so keep them short and testable.
### Requirements
- Observable behavior and constraints.
## 6. Interfaces and data
APIs, commands, events, schemas, config, compatibility, or migration.
### Naming and identity
How every stored name or ID is created, what happens when that fails, and what happens if its source changes after data exists.
## 7. Failure behavior and lifecycle
Say what can fail, what state follows, whether the system retries, and how it recovers. Cover startup, config or state changes, work in flight, shutdown, and what happens when several things fail together.
## 8. Security, privacy, and operations
State the trust boundary, authorization checks, sensitive data handling, and operational impact. Name shared limits such as rate limits, connections, disk, memory, or cost. Say what happens at each limit.
## 9. Acceptance criteria
- `AC-1`: Testable condition that proves the work is complete.
## 10. Test approach
How each `INV-n` and `AC-n` will be proved. Cite the IDs.
## 11. Risks and tradeoffs
- Risk and mitigation.
## 12. Open questions
- Question, and whether it blocks starting work.
## 13. Out of scope
- Related work this design does not include.
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 · 121 lines · 55 tokens per session scan A 226d254267e1
design is a skill published in the GitHub repository owainlewis/blueprint (378 stars, last pushed 7d ago), licensed MIT. It adds 55 tokens to every session and 1,465 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 2026-08-30.
Other skills, from other repositories
build-executor
Govern implementation from an approved execution contract. Invoke when execution-contract.md is approved and the user wants disciplined build work, TDD execution, or guarded batch-by-batch implementation.
workflow-start
Primary entry point for the spec-superflow state-machine workflow. Invoke when the user is inside an active spec-superflow change directory (look for .spec-superflow.yaml, changes/ /, proposal.md, specs/, design.md, tasks.md, or execution-contract.md) and asks to start, continue, resume, implement, plan, or figure out…
bug-investigator
Use when encountering any bug, test failure, or unexpected behavior during spec-superflow execution, before proposing fixes. Invoked automatically when build-executor hits a blockage.
release-archivist
Close out a spec-superflow change with verification, summary, and archive readiness. Invoke when implementation is complete, verification is underway, or the user asks for a final wrap-up.
contract-builder
Convert approved planning artifacts into an execution contract. Invoke when the user wants to start building, asks to move from planning to implementation, or when execution-contract.md is missing or stale.
need-explorer
Clarify intent, scope, constraints, and success criteria before artifact creation. Invoke when the request is fuzzy, the user is comparing options, or the workflow needs a stable change definition before writing artifacts.