New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Agent Native Architecture

by sandgardenhq in sandgardenhq/sgai

Build AI agents using prompt-native architecture where features are defined in prompts, not code.

Custom licenceAuto-check passedAgent Workflows

Install Agent Native Architecture

skills CLI
$ npx skills add sandgardenhq/sgai --skill agent-native-architecture -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install sandgardenhq/sgai agent-native-architecture --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/sandgardenhq/sgai.git skills-src && mkdir -p .claude/skills && cp -r skills-src/cmd/sgai/skel/.sgai/skills/architecture/agent-native-architecture .claude/skills/agent-native-architecture && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
agent-native-architecture
GitHub stars
137
Token cost
~2k tokens
SKILL.md length
677 words
Files
6 (incl. references)
Skills in repo
53
Repo updated
First seen
Licence
Custom licence

At a glance

Build AI agents using prompt-native architecture where features are defined in prompts, not code.

  • Works in 4 steps: Start in the prompt - New features begin… → Iterate rapidly - Change behavior by… → Graduate when stable - Harden to code… → …
  • Creating autonomous agents
  • SKILL.md covers The Prompt-Native Philosophy, Domain Knowledge and What NOT to Do
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Agent Native Architecture is an agent skill from sandgardenhq/sgai. Build AI agents using prompt-native architecture where features are defined in prompts, not code. Use when creating autonomous agents, designing MCP servers, implementing self-modifying systems, or adopting the "trust the agent's intelligence" philosophy.

Its SKILL.md is about 2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 6 other files, including reference files (for example `references/architecture-patterns.md`, `references/mcp-tool-design.md` and `references/refactoring-to-prompt-native.md`).

It sits in Agent Workflows, covering Autonomous loops and MCP servers. It works with Model Context Protocol. The repository describes itself as: Sandgarden AI Software Factory.

When your agent uses it

  • Creating autonomous agents
  • Designing MCP servers
  • Implementing self-modifying systems
  • Adopting the trust the agents intelligence philosophy

Example prompts

  • “trust the agent”
  • “/agent-native-architecture”

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Start in the prompt - New features begin as natural language defining outcomes
  2. Iterate rapidly - Change behavior by editing prose, not refactoring code
  3. Graduate when stable - Harden to code when requirements stabilize AND speed/reliability matter
  4. Many features stay as prompts - Not everything needs to become code

What it can do on your machine

Read from SKILL.md and the folder at commit 9efbb7b. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are typescript and markdown).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Agent Native Architecture loads about 2k tokens when it runs, and up to ~12k if it reads all its reference files. Until then it costs about 70 tokens; SKILL.md has 677 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~70
When it runs · the whole SKILL.md, loaded when a task matches
~2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~12k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 677 words (~1,999 tokens).

“Agent native engineering inverts traditional software architecture. Instead of writing code that the agent executes, you define outcomes in prompts and let the agent figure out HOW to achieve them.”

— opening of SKILL.md by sandgardenhq, Custom licence
name
agent-native-architecture

Read the full SKILL.md on GitHub

Files

SKILL.md and 5 other files (references) in cmd/sgai/skel/.sgai/skills/architecture/agent-native-architecture of sandgardenhq/sgai.

  • SKILL.md
  • references/architecture-patterns.md
  • references/mcp-tool-design.md
  • references/refactoring-to-prompt-native.md
  • references/self-modification.md
  • references/system-prompt-design.md

Open the folder on GitHubat commit 9efbb7b

Compare with similar skills

Agent Native Architecture next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Agent Native Architecture compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Agent Native Architecture this skillsandgardenhq/sgai137—~2kAutomated safety check: PassCustom licence
Ouroboros Evolve LoopQ00/ouroboros6.2k—~3.2kAutomated safety check: PassMIT
Hermes Mission ControlTh0rgal/sandboxed.sh516—~8.6kAutomated safety check: PassNone
Ouroboros Command ReferenceQ00/ouroboros6.2k—~1.8kAutomated safety check: PassMIT
Ouroboros Ralph LoopQ00/ouroboros6.2k—~3.5kAutomated safety check: PassMIT
Toolifycoreyhaines31/makerskills851—~2.9kAutomated safety check: NotesMIT

Similar skills

  • Ouroboros Evolve Loop

    Q00/ouroboros

    Starts, monitors or rewinds an evolutionary development loop that refines an ontology and acceptance criteria generation by generation until it converges, using the Ouroboros MCP tools.

    6.2k GitHub stars~3.2k tokensUpdated 3 days ago
    Agent WorkflowsAuto-check passed

  • Hermes Mission Control

    Th0rgal/sandboxed.sh

    Teaches Hermes to monitor and steer long-running sandboxed.sh missions: spot where a model is stuck, switch backends or models between turns, and send targeted hints.

    516 GitHub stars~8.6k tokensUpdated yesterday
    Agent WorkflowsAuto-check passed

  • Ouroboros Command Reference

    Q00/ouroboros

    Reference guide to the Ouroboros commands and agents, covering interviews, seed specs, evaluation, lateral-thinking personas and the evolutionary loop.

    6.2k GitHub stars~1.8k tokensUpdated 3 days ago
    Agent WorkflowsAuto-check passed

  • Ouroboros Ralph Loop

    Q00/ouroboros

    Runs a persistent Ouroboros loop that repeats background evolve_step generations on a lineage until QA passes, evolution converges, or you cancel.

    6.2k GitHub stars~3.5k tokensUpdated 3 days ago
    Agent WorkflowsAuto-check passed

  • Toolify

    coreyhaines31/makerskills

    When you want to integrate an external tool, API, MCP server, or service into a project — the wizard walks you through auth, config, env vars, client wrapper code, example usage, and an optional…

    851 GitHub stars~2.9k tokensUpdated 2 days ago
    Agent WorkflowsAuto-check: notes

  • Improving MCP Tools

    PostHog/posthog

    Official

    Run an improve-my-MCP campaign: an autoresearch-style loop that measures the MCP agent experience with the eval harness, picks the highest-impact tool problem from production data, makes one bounded…

    40k GitHub stars~1.5k tokensUpdated today
    Agent WorkflowsAuto-check passed

More from sandgardenhq/sgai

All 53 skills in this repo

  • Root Cause Tracing

    sandgardenhq/sgai

    A skill your agent uses when errors occur deep in execution and you need to trace back to find the original trigger - systematically traces bugs backward through call stack, adding instrumentation…

    137 GitHub starsUsed in 4 repos~1.4k tokens
    Auto-check passed

  • Condition Based Waiting

    sandgardenhq/sgai

    A skill your agent uses when tests have race conditions, timing dependencies, or inconsistent pass/fail behavior - replaces arbitrary timeouts with condition polling to wait for actual state…

    137 GitHub starsUsed in 3 repos~933 tokens
    Auto-check passed

  • Defense In Depth

    sandgardenhq/sgai

    A skill your agent uses when invalid data causes failures deep in execution, requiring validation at multiple system layers - validates at every layer data passes through to make bugs structurally…

    137 GitHub starsUsed in 3 repos~970 tokens
    Auto-check passed

  • Adhoc

    sandgardenhq/sgai

    Run and manage ad-hoc AI prompts in sgai workspaces without starting a full agentic session.

    137 GitHub stars~1.2k tokensUpdated 20 days ago
    Auto-check passed

  • Browser Bug Testing Workflow

    sandgardenhq/sgai

    You must use this skill when debugging web UI bugs or testing interactive components that require multi-step browser interactions.

    137 GitHub stars~3.7k tokensUpdated 20 days ago
    Auto-check passed

  • Create Opencode Agents

    sandgardenhq/sgai

    Guide for creating custom opencode agents with proper configuration, permissions, path-based delegation, work splitting, and strict status envelopes.

    137 GitHub stars~1.8k tokensUpdated 20 days ago
    Auto-check passed

Works with

Categories

Questions about Agent Native Architecture

What does Agent Native Architecture do?

Build AI agents using prompt-native architecture where features are defined in prompts, not code. Agent Native Architecture is an agent skill from sandgardenhq/sgai. Build AI agents using prompt-native architecture where features are defined in prompts, not code.

When should I use Agent Native Architecture?

Agent Native Architecture fits situations like: creating autonomous agents; designing MCP servers; implementing self-modifying systems; adopting the trust the agents intelligence philosophy.

How do I install Agent Native Architecture in Claude Code?

Run `npx skills add sandgardenhq/sgai --skill agent-native-architecture -a claude-code`. Or copy the skill folder (cmd/sgai/skel/.sgai/skills/architecture/agent-native-architecture in sandgardenhq/sgai) into .claude/skills/agent-native-architecture in your project. Claude Code loads it when a task matches its description.

How do I install Agent Native Architecture in Codex?

Run `npx skills add sandgardenhq/sgai --skill agent-native-architecture -a codex`. Or copy the skill folder (cmd/sgai/skel/.sgai/skills/architecture/agent-native-architecture in sandgardenhq/sgai) into .agents/skills/agent-native-architecture in your project. Codex loads it when a task matches its description.

Can I use Agent Native Architecture in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add sandgardenhq/sgai --skill agent-native-architecture -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/agent-native-architecture, .gemini/skills/agent-native-architecture, .github/skills/agent-native-architecture and .opencode/skills/agent-native-architecture in your project.

What does Agent Native Architecture need to run?

SKILL.md names no scripts, command-line tools or credentials: Agent Native Architecture is instructions for the agent only.

Does Agent Native Architecture access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Agent Native Architecture safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Agent Native Architecture use?

Agent Native Architecture has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.

How many tokens does Agent Native Architecture use?

About 2k tokens (SKILL.md is roughly 8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 9.9k tokens, read only when the agent opens those files.

What are the alternatives to Agent Native Architecture?

Skills that share tags, products or a category with Agent Native Architecture: Ouroboros Evolve Loop (Q00/ouroboros, 6.2k stars), Hermes Mission Control (Th0rgal/sandboxed.sh, 516 stars), Ouroboros Command Reference (Q00/ouroboros, 6.2k stars) and Ouroboros Ralph Loop (Q00/ouroboros, 6.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Agent Native Architecture?

sandgardenhq (a GitHub organization) maintains it in sandgardenhq/sgai, which has 137 GitHub stars. The repository holds 53 skills in this directory. The repository was last updated on September 21, 2026.

Source: sandgardenhq/sgai on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.