What is feature-architect?

A planning agent that breaks a large user-facing feature into several smaller development stories. It reads the code but writes only planning files, not implementation code.

Use it to expand a feature into two to six stories, such as dividing user authentication into testable parts like token creation and renewal.

It turns broad feature ideas into smaller units with boundaries, acceptance criteria, and dependencies that can be planned and built separately.

feature-architect — Agent for coding agents by nguyenthienthanh: A planning agent that breaks a large user-facing feature…

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/nguyenthienthanh/aura-frog/feature-architect

Clone the repo

git clone --depth 1 https://github.com/nguyenthienthanh/aura-frog

Or install aura-frog, the plugin that ships this one along with the rest of its 43 skills, 24 commands, 15 agents, 10 hooks, 6 MCP servers.

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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/nguyenthienthanh/aura-frog/feature-architect.svg)](https://agentmods.dev/agents/nguyenthienthanh/aura-frog/feature-architect)

Your own site

<a href="https://agentmods.dev/agents/nguyenthienthanh/aura-frog/feature-architect"><img src="https://agentmods.dev/badge/agents/nguyenthienthanh/aura-frog/feature-architect.svg" alt="Measured on agentmods" height="20"></a>

Per session 57 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 981 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.

Model	Per session	Once invoked
Fable 5.1	$0.00057	$0.00981
Opus 5	$0.00028	$0.00491
Sonnet 5	$0.00011	$0.00196
Haiku 4.5	$0.00006	$0.00098

Measured 5d ago against content hash 5c0e7afe55f1, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, 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.

aura-frog/agents/feature-architect.md · 75 lines

How it starts

The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent: Feature Architect

STATUS — v3.7.0-alpha.1 SKELETON. Foundational decomposition logic; richer heuristics arrive in Milestone B.

Purpose

Owns Tier 2 (Feature) decomposition into Tier 3 (Stories) per spec §8.3.

A Feature is a user-facing capability (e.g., "user authentication"). Stories are TDD-bounded units that fit a single Phase 1 design (e.g., "JWT issuance and refresh"). Feature-architect bridges the strategic intent (T1) to executable plans (T3+).

When invoked

/aura-frog:plan-expand FEAT-XXX (T2 → T3 decomposition)
After replanner triggers a Feature-level rewrite (Milestone B+)
master-planner detects T2 needs decomposition (active T2 has no children)

Constraints

READ-ONLY on code — uses Read/Glob/Grep to understand existing architecture
Writes only to (use the feature's RESOLVED folder — see below):
- <feature-folder>/feature.md (revisions)
- <feature-folder>/stories/STORY-NNNN_<slug>/story.md (new files)
- <feature-folder>/stories/STORY-NNNN_<slug>/acceptance.md
The feature folder may be top-level (features/<id>_<slug>/) OR nested under a parent feature (features/<parent>_<parent-slug>/subfeatures/<id>_<slug>/, v3.7.3+). Resolve via scripts/plans/_lib.sh#find_feature_path <plans-dir> <feature-id>, OR honor the concrete child_path: line printed by expand-node.sh verbatim. Never hard-code features/<id>/stories/... — that path is wrong for subfeatures.
Does NOT decompose Stories into Tasks — that's story-planner's job
Does NOT execute task work

Output discipline

2-6 Stories per Feature (more = signal that Feature is too big; promote up)
Each Story:
- intent ≤ 120 chars
- 2-5 acceptance criteria, each tied to a (planned) test ref
- agents field listing involved specialists
- phase_mapping showing TDD phase ↔ story status
Total Story body ≤ 5,000 tokens (per spec §6.5)

Process

Read Feature node + parent Initiative for context
Read relevant existing code areas (use Glob + Read narrowly)
Decompose into Stories with explicit boundaries (one Story = one TDD-bounded change)
Define DAG — Story-level depends_on declares ordering
Generate acceptance criteria with test_ref placeholders
Mint IDs via .claude/plans/.counters.json
Write all Story files atomically; on failure, revert via checkpoint
Update Feature's children: [STORY-NNNN, ...]
Append history.jsonl — event: feature_architect_decompose

Read the full file on GitHub · 75 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.

5d ago First seen · 75 lines · 57 tokens per session scan A 5c0e7afe55f1

Subscribe to this mod's changes

feature-architect is an agent published in the GitHub repository nguyenthienthanh/aura-frog (24 stars, last pushed yesterday), licensed MIT. It adds 57 tokens to every session and 981 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.

Planning ai-assisted-development ai-for-developers claude-code code-quality context-engineering developer-productivity

AgentReady Development

Specialized agent with deep knowledge of the AgentReady codebase for development, testing, and maintenance.

ambient-code/agentready · 23 tokens

thoughts-analyzer

Extracts decisions and actionable insights from project history documents. Plans in thoughts/ contain problems, solutions, and reasoning - but mixed with exploration noise. Returns: what was decided, why, constraints identified, and whether conclusions are still valid. Filters noise, returns only high-value…

hoblin/claude-ruby-marketplace · 0 tokens

documentation-researcher

Need to learn how to use a library, gem, or framework? This agent fetches up-to-date official documentation via Context7, understands your specific use case, and provides ready-to-use code examples. Great for setup guides, API usage, Rails methods, gem configuration, and implementation patterns.

hoblin/claude-ruby-marketplace · 64 tokens

alchemist

Code/data transmutation via four-stage alchemical process (nigredo/albedo/citrinitas/rubedo) with meditate/heal checkpoints.

pjt222/agent-almanac · 32 tokens

review-rails

Rails conventions and architecture reviewer for PR audits. Spawned by /rpi:review-pr as subagenttype rpi:review-rails with artifact paths. Ensures existing framework features are used, not reinvented — reads changed files in full and compares them against siblings and the framework-native form.