architect

A command that delegates the creation of an architecture decision record, technical proposal, or agreement between services to a specialist agent.

In plain words
What is it for?
Use it to write an ADR, RFC, or cross-service contract, optionally using context from a referenced ticket.
Why use it?
It helps document important technical choices and shared interfaces before implementation begins.

Command

Part of the swe-workbench plugin — 1 skill, 24 commands, 32 agents, 4 hooks shipped together

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 commands/lugassawan/swe-workbench/architect
Clone the repo
git clone --depth 1 https://github.com/lugassawan/swe-workbench

Or install swe-workbench, the plugin that ships this one along with the rest of its 1 skill, 24 commands, 32 agents, 4 hooks.

Per session 15 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 722 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 $0.00015 $0.00722
Opus 5 $0.00008 $0.00361
Sonnet 5 $0.00003 $0.00144
Haiku 4.5 $0.00002 $0.00072

Measured 2d ago against content hash f9038696c4dd, 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.

commands/architect.md · 29 lines

What it actually says

Decision: $ARGUMENTS

If $ARGUMENTS contains a ticket reference, invoke swe-workbench:ticket-context first and prepend its structured summary to the delegation context below. Skip if $ARGUMENTS is free-text with no recognizable ref. (Trigger patterns are defined in that skill's "When to invoke" section.)

Interrogation mode. Before producing anything, resolve the mode:

  • Explicit signal in the invocation is honored without asking. grill-me = --grill, "grill me", or "grill-me mode". standard = --standard, "standard", or "quick". Strip the signal from $ARGUMENTS and record the resolved mode.
  • No explicit signal: ask via AskUserQuestion — one question, header "Mode", options Standard (recommended, listed first) and Grill me. Standard description: "Lightweight clarify — a restatement and at most one question, then proceed." Grill-me description: "Relentlessly walk the decision tree one question at a time, each with a recommended answer, self-answering from the codebase where possible." Use the user's choice.

Standard mode: proceed with the command's existing lightweight clarify (a restatement and at most one clarifying question) — do not ask the mode question again.

Grill-me mode: activate swe-workbench:workflow-grill and run its interrogation loop to completion (exit on shared understanding or when the user says "proceed"). Then thread the emitted ## Resolved decisions block into the command's normal artifact/delegation step below — the same way a ticket-context summary is prepended — and continue as in standard mode.

Delegate to the swe-workbench:architect subagent. Its output must include:

  1. Decision — one paragraph stating what was chosen and why.
  2. Context & forcing function — why this decision is needed now.
  3. Options considered — table or bullets; all four sub-fields required for each option: strengths, weaknesses, operational cost, reversibility classification (one-way / two-way door).
  4. Consequences — positive and negative; what becomes easier and what becomes harder or more expensive.
  5. Open questions — what remains undecided and why it is safe to defer.
  6. References — RFC numbers, prior ADRs, principle skills consulted, external standards cited.

Plan output: If you (the orchestrator) author a plan based on the subagent's response and that plan modifies the codebase (fix / make / implement) — whether saved to a plan file or passed to ExitPlanMode — first activate swe-workbench:workflow-development in Mode A and embed the rendered ## Workflow section in the plan per skills/workflow-development/templates/plan-workflow-section.md. Run the skill's project-detection (git branch -a, git log --oneline -20, Makefile grep, PR-template lookup) so the placeholders are substituted from this repo, not left as [[detect:…]]. Skip Mode A if the plan is pure analysis, design recommendation, or any output that does not introduce file edits — the Workflow section's phases (Branch / Verify / Deliver) do not apply.

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 · 29 lines · 15 tokens per session scan A f9038696c4dd

Subscribe to this mod's changes

architect is a command published in the GitHub repository lugassawan/swe-workbench (2 stars, last pushed 2d ago), licensed MIT. It adds 15 tokens to every session and 722 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.