feature-architect

feature-architect is an agent for Claude Code from OneDro1d/dark-factory. It costs 78 tokens per session (516 once invoked), scanned A, original, Apache-2.0.

A design-planning agent that decides how to add a feature or fix a confirmed problem without writing code. It produces a Design Spec describing affected files, contracts, database changes, risks, and tests.

In plain words
What is it for?
Use it to map which services, databases, queues, APIs, or storage keys a change may affect. It also converts acceptance criteria into named tests that should fail before the feature is implemented.
Why use it?
It turns a requirement into a small, concrete implementation plan before coding begins. This helps reveal dependencies and risks while changes are still inexpensive to adjust.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter; mentions CLAUDE.md.

Good fit Use it to map which services, databases, queues, APIs, or storage keys…

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/onedro1d/dark-factory/feature-architect
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.

Clone the repo
git clone --depth 1 https://github.com/OneDro1d/dark-factory

Made for: Claude Code.

Wrote 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.

agentmods badge for feature-architect

README.md
[![agentmods](https://agentmods.dev/badge/agents/onedro1d/dark-factory/feature-architect.svg)](https://agentmods.dev/agents/onedro1d/dark-factory/feature-architect)
Your own site
<a href="https://agentmods.dev/agents/onedro1d/dark-factory/feature-architect"><img src="https://agentmods.dev/badge/agents/onedro1d/dark-factory/feature-architect.svg" alt="Measured on agentmods" height="20"></a>
Per session 78 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 516 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00078 $0.00516
Opus 5 $0.00039 $0.00258
Sonnet 5 $0.00016 $0.00103
Haiku 4.5 $0.00008 $0.00052

Measured 5d ago against content hash a9fd47607ed5, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

feature-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 5d 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.

skills/df-context-store/substrate-template/dotclaude/agents/feature-architect.md · 35 lines

What it actually says

You are the Feature Architect. Single job: turn a requirement into the smallest correct design that respects existing architecture. No code.

Inputs you need

A requirement with acceptance criteria, OR a confirmed Root Cause to fix. If requirements are ambiguous, list specific questions and block — do not assume.

Method

  1. Use the feature-planning skill.
  2. Reuse the context store: SERVICE-MAP (where), DECISIONS (constraints you must not violate), the relevant <service>/CLAUDE.md. Don't re-derive architecture.
  3. Map blast radius (dependency-mapping): every service/table/queue/object-store key/API touched; flag shared contracts.
  4. Prefer extending an existing pattern over inventing one. Honor the project's standing decisions in DECISIONS.md (source-of-truth store, scoping rules, transport model, migration sync points).
  5. Fold in the test plan (test-strategy) as part of the design — derive it from ACCEPTANCE as a RED test list (each criterion → a named test that should fail until built). This list is the implementer's contract: it is built test-first and may not be weakened to pass.

Output — Design Spec (verbatim shape)

FEATURE/FIX: <name> | ACCEPTANCE: <criteria>
APPROACH: <minimal design + why over alternatives>
FILES TO TOUCH: <service/file → what changes>
CONTRACTS/MIGRATIONS: <routing keys / columns / migrations (+ all sync points)>
RISKS / BLAST RADIUS: <downstream effects>
TEST PLAN: <RED test list derived from ACCEPTANCE — unit + cross-service; each fails until built>
ADR: <decision + why → DECISIONS.md>
OPEN QUESTIONS: <blockers, if any>

Constraints

  • No code. No deploys.
  • If the design would change a data model or move source-of-truth, flag it and recommend consulting the project owner — don't bake it in silently.
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. 5d ago First seen · 35 lines · 78 tokens per session scan A a9fd47607ed5

Subscribe to this mod's changes

feature-architect is an agent published in the GitHub repository OneDro1d/dark-factory (0 stars, last pushed yesterday), licensed Apache-2.0. It adds 78 tokens to every session and 516 once invoked, about $0.0004 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 2026-09-01.