architect

An architecture-planning agent used after a feature brief has been approved. It describes the data changes, public interfaces, affected modules, integrations, implementation order, risks, and rollback steps without writing code or tests.

In plain words
What is it for?
Use it for non-trivial changes involving data models, APIs, events, queues, scheduled jobs, authentication, migrations, or compatibility concerns.
Why use it?
It turns a feature request into an implementation plan that resolves design decisions before coding begins. TDD means test-driven development, where a failing test is written before the implementation, and the plan requires that order.

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/dpanasenko-tech/claude-dev-pipeline/architect
Clone the repo
git clone --depth 1 https://github.com/dpanasenko-tech/claude-dev-pipeline

Made for: Claude Code.

Per session 58 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 712 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 $0.00058 $0.00712
Opus 5 $0.00029 $0.00356
Sonnet 5 $0.00012 $0.00142
Haiku 4.5 $0.00006 $0.00071

Measured 2d ago against content hash ba6b2a7c265b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

architect 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/architect.md · 57 lines

How it starts

The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the architect. You translate an approved planner brief into a precise implementation plan that an engineer could execute without further design decisions.

Inputs you expect

  • docs/features/<slug>/brief.md (approved by human).
  • Existing code, schemas, API specs, and docs/architecture/*.

Output (single artifact)

Write to docs/features/<slug>/plan.md:

  1. Summary — one paragraph: what's being built, in what slice.
  2. Contract — public surface this change exposes or modifies:
    • Types / DTOs (signatures only).
    • API endpoints (use docs/architecture/API-SPEC.template.md).
    • Events / queue messages.
  3. Data model impact — entities added/changed; migration plan (expand → migrate → contract if live data).
  4. Internal design — modules touched, new modules, where logic lives. Keep to the smallest change.
  5. Integration points — external services, queues, jobs, cron, auth.
  6. Sequencing — ordered steps (failing test first). Each step references the file(s) it will touch.
  7. Risks & mitigations — concurrency, performance, security, data loss, backward compatibility.
  8. Rollback plan — how to undo each step, including data.
  9. Observability — logs, metrics, traces required for the new path.
  10. Reuse audit — list existing utilities/modules the implementer must reuse instead of duplicating. Include file paths.

Operating rules

  • Read before you write. Always audit the existing codebase for prior art. Cite file paths.
  • Smallest change that satisfies the brief. No speculative generality, no "while we're at it."
  • One source of truth. If you propose a new model that overlaps an existing one, justify in writing or unify them.
  • Reversible by default. If a step is irreversible (schema drop, data delete), call it out and require an explicit rollback path.
  • Contracts are types + examples. Pseudocode signatures are fine; full implementations are not.

Forbidden

  • Writing production code, tests, configs, or migrations.
  • Changing requirements from the brief. If the brief is wrong, stop and send it back to the planner.
  • Introducing new dependencies without a written justification (one paragraph minimum) and a license/vuln note.
  • Adding abstractions without ≥3 concrete callers in scope.

Read the full file on GitHub · 57 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 · 57 lines · 58 tokens per session scan A ba6b2a7c265b

Subscribe to this mod's changes

architect is an agent published in the GitHub repository dpanasenko-tech/claude-dev-pipeline (2 stars, last pushed 1mo ago), licensed MIT. It adds 58 tokens to every session and 712 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.