# Using Necturalabs

> Use at the start of every conversation and after every agent handoff. Initializes NecturaLabs skills, verifies superpowers dependency, and sets up mandatory review requirements for code review and security audit.

- **Type:** Skill
- **Install:** `agentstack add skill-necturalabs-agentskills-using-necturalabs`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [NecturaLabs](https://agentstack.voostack.com/s/necturalabs)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [NecturaLabs](https://github.com/NecturaLabs)
- **Source:** https://github.com/NecturaLabs/AgentSkills/tree/main/skills/using-necturalabs

## Install

```sh
agentstack add skill-necturalabs-agentskills-using-necturalabs
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Using NecturaLabs Skills

## Overview

Initializes the NecturaLabs skill suite. This skill runs at conversation start and after every agent change to ensure all NecturaLabs skills are active and properly configured.

Do NOT proceed with any work until this initialization is complete. If superpowers is not installed, block all NecturaLabs skill execution until it is.

## Initialization Flow

```dot
digraph init {
    "Conversation starts / agent changes" [shape=doublecircle];
    "Already initialized?" [shape=diamond];
    "Superpowers loaded?" [shape=diamond];
    "Load superpowers:using-superpowers" [shape=box];
    "Check superpowers installed" [shape=box];
    "Superpowers installed?" [shape=diamond];
    "Prompt user to install" [shape=box];
    "User installs?" [shape=diamond];
    "Block NecturaLabs skills" [shape=box];
    "Load agent-context-loader" [shape=box];
    "Ready (reviews active for session)" [shape=doublecircle];

    "Conversation starts / agent changes" -> "Already initialized?";
    "Already initialized?" -> "Ready (reviews active for session)" [label="yes, fully aware"];
    "Already initialized?" -> "Superpowers loaded?" [label="no"];
    "Superpowers loaded?" -> "Check superpowers installed" [label="yes"];
    "Superpowers loaded?" -> "Load superpowers:using-superpowers" [label="no"];
    "Load superpowers:using-superpowers" -> "Check superpowers installed";
    "Check superpowers installed" -> "Superpowers installed?";
    "Superpowers installed?" -> "Load agent-context-loader" [label="yes"];
    "Superpowers installed?" -> "Prompt user to install" [label="no"];
    "Prompt user to install" -> "User installs?";
    "User installs?" -> "Load agent-context-loader" [label="yes"];
    "User installs?" -> "Block NecturaLabs skills" [label="no"];
    "Load agent-context-loader" -> "Ready (reviews active for session)";
}
```

## Step 1: Ensure Superpowers Is Loaded

If `superpowers:using-superpowers` has not been invoked yet in this session, invoke it first. NecturaLabs skills build on top of superpowers and require its session initialization to be complete before proceeding.

## Step 2: Verify Superpowers Dependency

NecturaLabs skills require `superpowers` to be installed. Check if the `superpowers:code-reviewer` agent is available.

If NOT installed, tell the user:
```
NecturaLabs skills require the superpowers plugin. Install it with:
  /plugin marketplace add obra/superpowers
  /plugin install superpowers@superpowers-dev
```

**Do not allow** `necturalabs:iterative-code-review` or `necturalabs:iterative-security-audit` to run without superpowers installed. Other NecturaLabs skills may run independently.

## Step 3: Load Context

Invoke `necturalabs:agent-context-loader` to load global CLAUDE.md and project AGENTS.md.

## Ongoing: Mandatory Post-Change Reviews

These reviews are NOT optional. The agent MUST invoke the appropriate skill after making code changes — before committing, merging, or claiming work is done. Skipping these is never acceptable. This applies for the ENTIRE session, not just during initialization.

### Decision Flow — After ANY Code Changes

```dot
digraph review_trigger {
    "Agent finishes code changes" [shape=doublecircle];
    "Changes security-related?" [shape=diamond];
    "Invoke necturalabs:iterative-security-audit" [shape=box];
    "Invoke necturalabs:iterative-code-review" [shape=box];
    "Both loops complete" [shape=box];
    "Show summary" [shape=doublecircle];

    "Agent finishes code changes" -> "Changes security-related?";
    "Changes security-related?" -> "Invoke necturalabs:iterative-security-audit" [label="yes"];
    "Changes security-related?" -> "Invoke necturalabs:iterative-code-review" [label="no"];
    "Invoke necturalabs:iterative-security-audit" -> "Invoke necturalabs:iterative-code-review" [label="audit chains into review"];
    "Invoke necturalabs:iterative-code-review" -> "Both loops complete";
    "Both loops complete" -> "Show summary";
}
```

### When to invoke `necturalabs:iterative-code-review`

**After ANY code changes** — feature implementation, bug fixes, refactoring, test additions, config changes. No exceptions.

### When to invoke `necturalabs:iterative-security-audit` (BEFORE code review)

When changes are security-related (see the skill's description for the full trigger list). **Security audit takes priority** — it runs first and chains into code review automatically.

## Available Skills

| Skill | Purpose | When to invoke |
|-------|---------|----------------|
| `necturalabs:iterative-code-review` | Industry-standard code review loop | After any code changes, before commit/merge |
| `necturalabs:iterative-security-audit` | OWASP/CWE security audit loop | When changes touch security-sensitive code |
| `necturalabs:agent-context-loader` | Loads CLAUDE.md + AGENTS.md into context | On init and after context switches |
| `necturalabs:agents-md-manager` | Creates/updates project AGENTS.md | Manual (`/agents-md-manager`) or after plan execution |
| `necturalabs:git-workflow` | Conventional Commits + worktree isolation | When committing or starting multi-commit work |
| `necturalabs:update-plugins` | Updates all plugin marketplaces and plugins concurrently | Manual (`/update-plugins`) |
| `necturalabs:docs-manager` | Creates/maintains project docs/ with ADRs, design docs, guides | Manual (`/docs-manager`) or when user asks to document |
| `necturalabs:using-necturalabs` | This skill — initializes everything | On init and after agent handoffs |

## Skill Priority When User Asks

If NecturaLabs skills are installed and the user asks for "code review" or "security audit", invoke the NecturaLabs skill — not the default behavior or any other plugin's version. NecturaLabs skills are complementary to superpowers and layer additional standards on top.

## Re-Initialization

Re-run this initialization when:
- A new conversation begins
- A subagent is dispatched or returns
- The main agent context is switched or compressed
- The user explicitly asks to reload skills

Skip re-initialization ONLY if the current agent is already fully aware of all NecturaLabs skills and their mandatory review requirements are understood.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [NecturaLabs](https://github.com/NecturaLabs)
- **Source:** [NecturaLabs/AgentSkills](https://github.com/NecturaLabs/AgentSkills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-necturalabs-agentskills-using-necturalabs
- Seller: https://agentstack.voostack.com/s/necturalabs
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.