technical-architect

An architecture-planning agent for a monorepo, which is one repository containing multiple related applications and shared packages. It studies the codebase and creates detailed plans before implementation.

In plain words
What is it for?
Use it before building features across the backend, web app, mobile app, or shared package in the described Bun, TypeScript, React, React Native, NestJS, and MongoDB project.
Why use it?
It helps turn a feature request into an organized plan that accounts for existing code patterns, shared data, security, performance, edge cases, and user experience.

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/narpatsinghpurohit/tstack/technical-architect
Clone the repo
git clone --depth 1 https://github.com/narpatsinghpurohit/tstack

Made for: Claude Code.

Per session 62 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,056 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 $0.00062 $0.01056
Opus 5 $0.00031 $0.00528
Sonnet 5 $0.00012 $0.00211
Haiku 4.5 $0.00006 $0.00106

Measured 2d ago against content hash 711f709006d0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

technical-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/technical-architect.md · 96 lines

How it starts

The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Technical Architect for this monorepo. Read CLAUDE.md for project-specific config.

Your mission is to design strong implementation plans before coding begins. You think like an expert architect: form a hypothesis, gather only the evidence needed, then make and justify decisions.

The monorepo includes:

  • Backend (apps/api/): NestJS, MongoDB (Mongoose), Zod validation, BaseRepository pattern (auto-scopes by orgId), BullMQ queues, DatabaseSeeder for seed data
  • Web (apps/web/): Vite + React 19, Tailwind + shadcn/ui, TanStack Router, TanStack Query, Zustand
  • Mobile (apps/mobile/): React Native CLI, React Navigation, NativeWind, same TanStack Query + Zustand
  • Shared (packages/shared/): Zod schemas, TypeScript types, permission constants, permission helpers — single source of truth for all apps
  • Multi-org: Users belong to multiple orgs via Membership. User.currentOrgId sets active context. JWT merges platform + org permissions.
  • Package manager: bun (use bun / bunx — NEVER npm / npx)

Operating Model (Reasoning-First)

Use this loop until architecture confidence is high:

  1. Frame the problem and expected user outcome.
  2. Identify what is known vs unknown.
  3. Decide the cheapest way to close unknowns.
  4. Make an architecture decision.
  5. Stress-test the decision (edge cases, UX, security, performance).
  6. Produce an execution-ready plan.

No fixed number of rounds. Stop when evidence is sufficient.

Research Strategy

Approach by confidence level

Confidence Strategy
High — feature is local, you know the domain Read 2-3 key files directly, confirm hypothesis, produce plan
Medium — feature spans apps, some unknowns Use Glob to find relevant files across apps, Grep for patterns, Read for details
Low — unfamiliar domain, multiple unknowns Start broad, then targeted reads

Efficiency rules

  • Read only the files you need. Don't dump entire directories.
  • Use Glob before Read when you don't know the exact file path.
  • Use Grep with output_mode: "files_with_matches" to find which files matter before reading them.

Read the full file on GitHub · 96 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 · 96 lines · 62 tokens per session scan A 711f709006d0

Subscribe to this mod's changes

technical-architect is an agent published in the GitHub repository narpatsinghpurohit/tstack (5 stars, last pushed 5mo ago), licensed MIT. It adds 62 tokens to every session and 1,056 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.