agentic-architecture

agentic-architecture is a skill for Claude Code from jeet129/praxis. It costs 162 tokens per session (3,420 once invoked), scanned A, original, MIT.

A design guide for software features that use large language models (LLMs), including how many agents to use, how they use tools, and how they store and retrieve memory.

In plain words
What is it for?
Designing single- or multi-agent systems, planner-executor workflows, tool schemas, memory systems, structured outputs, streaming, and safeguards around LLM calls.
Why use it?
It helps separate predictable program logic from model-generated decisions and plan what happens when a model is wrong, slow, or unavailable. It also makes tool access and failure handling explicit.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions subagents.

Part of the praxis plugin — 105 skills, 12 commands, 17 agents, 6 hooks shipped together

Good fit Designing single- or multi-agent systems, planner-executor workflows, tool schemas, memory systems, structured outputs, streaming, and safeguards around LLM calls.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/jeet129/praxis/agentic-architecture
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.

Any agent
npx skills add jeet129/praxis --skill agentic-architecture
Clone the repo
git clone --depth 1 https://github.com/jeet129/praxis

Made for: Claude Code.

Or install praxis, the plugin that ships this one along with the rest of its 105 skills, 12 commands, 17 agents, 6 hooks.

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 agentic-architecture

README.md
[![agentmods](https://agentmods.dev/badge/skills/jeet129/praxis/agentic-architecture.svg)](https://agentmods.dev/skills/jeet129/praxis/agentic-architecture)
Your own site
<a href="https://agentmods.dev/skills/jeet129/praxis/agentic-architecture"><img src="https://agentmods.dev/badge/skills/jeet129/praxis/agentic-architecture.svg" alt="Measured on agentmods" height="20"></a>
Per session 162 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,420 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. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high Prompt Injection · line 8
    Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.
    Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
How audits are shown
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 $0.00162 $0.03420
Opus 5 $0.00081 $0.01710
Sonnet 5 $0.00032 $0.00684
Haiku 4.5 $0.00016 $0.00342

Measured 8d ago against content hash 0aec2ad6ced3, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

agentic-architecture 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 8d 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.

plugins/praxis-codex/skills/agentic-architecture/SKILL.md · 274 lines

How it starts

The opening of the file, as written. The whole thing — 274 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agentic Architecture

capability: agentic-ai
domain: ml
state: active
dependencies:
  - architecture-pattern-selection
  - resilience-patterns
  - distributed-systems-patterns
  - nfr-definition
  - ml-problem-framing
triggers:
  - "designing a new LLM-powered feature"
  - "choosing agent topology (single / multi / planner-executor / supervisor)"
  - "designing tool-use surfaces and tool schemas"
  - "deciding memory architecture (short-term / episodic / semantic)"
  - "placing LLM calls relative to deterministic logic"
  - "choosing structured-output / function-calling / streaming patterns"
outputs:
  - agent topology decision + ADR
  - tool catalog (per tool: schema, side effects, idempotency, authn/authz)
  - memory architecture (which memory types; storage; retrieval)
  - LLM-vs-deterministic boundary diagram (call graph)
  - structured-output / function-calling design
  - failure-mode catalog (what does the system do when the LLM is wrong / slow / unavailable)
consumers:
  - ml-ai-engineer (primary author)
  - solution-architect (co-designs system architecture)
  - rag-design (consumes when retrieval-augmented)
  - llm-safety (consumes for guardrail placement)
  - llm-cost-optimization (consumes for cost-routing decisions)
  - evaluation-engineering (consumes for what to evaluate)
references: []

The macro-design layer for LLM-powered features. Done right, it sets up rag-design, evaluation-engineering, llm-safety, and llm-cost-optimization to do their jobs. Done poorly, the system is a chatbot wrapped around a single API call with no story for failure, no story for evaluation, and no story for cost.

The principle: LLMs are non-deterministic components within a deterministic system. Architect the system; the LLM is one (important) part.

When this skill fires

  • A new LLM-powered feature is being designed.
  • Agent topology is being chosen (single-agent, multi-agent, planner-executor, supervisor).
  • Tool-use surfaces are being designed (what the agent can call).
  • Memory architecture is being chosen.
  • The boundary between LLM calls and deterministic logic is being placed.
  • Structured-output / function-calling patterns are being decided.

Read the full file on GitHub · 274 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. 8d ago First seen · 274 lines · 162 tokens per session scan A 0aec2ad6ced3

Subscribe to this mod's changes

agentic-architecture is a skill published in the GitHub repository jeet129/praxis (7 stars, last pushed 4d ago), licensed MIT. It adds 162 tokens to every session and 3,420 once invoked, about $0.0008 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.