Skip to content
Back to skills

Navigate

ASecurity

Routes ambiguous mechanism comparisons to the smallest appropriate agent mechanism and skill. Use when the user is unsure whether they need instructions, a prompt, skill, subagent, hook, scheduler, MCP, plugin, Claude workflow, or Codex workflow; if the artifact is already selected, its specialist skill wins.

  • 5 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added September 4, 2026
ai-agentsapi

Works with

  • claude code
  • api
  • mcp

Security analysis

A100/100

Pro scans all 3 files and shows the line behind each finding

Scanned September 4, 2026

npx -y skills add alessiomarcone/agent-fieldbook --skill navigate --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Navigate?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Navigate
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/alessiomarcone-navigate/badge)](https://www.skillsdirectory.com/skills/alessiomarcone-navigate)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: navigate
description: Routes ambiguous mechanism comparisons to the smallest appropriate agent mechanism and skill. Use when the user is unsure whether they need instructions, a prompt, skill, subagent, hook, scheduler, MCP, plugin, Claude workflow, or Codex workflow; if the artifact is already selected, its specialist skill wins.
---

# Navigate

Route the request. Do not absorb specialist work that belongs to a narrower skill.

## Reference

Read `references/knowledge-base.md` when the mechanism boundary or product branch is not obvious.

## Routing table

| Intent | Route |
|---|---|
| Improve or design a request | `prompt` |
| Fact-check or stress-test an output | `verify` |
| Decide, design, or configure delegated agents | `subagent` |
| Create or refine an agent skill | `skill` |
| Package or distribute reusable capabilities | `plugin` |
| Design a Cowork workflow | `cowork` |
| Operate a Codex-specific surface | `codex` |
| Build an official learning path | `learn` |
| Distill source material | `knowledge-card` |
| Run work on a clock or recurring schedule | `cowork` for Claude/Cowork, `codex` for Codex, or the host scheduler/API |

If no specialist matches, select the smallest supported mechanism: normal chat/task, project knowledge, `CLAUDE.md` or `AGENTS.md`, skill, subagent, hook, scheduler/automation, MCP/connector, plugin, or API/SDK. Hooks are event-driven; they are not a substitute for clock-driven scheduling.

## Output contract

Return three fields, plus an optional fourth:

1. **Route** — one skill or mechanism.
2. **Why** — one short paragraph.
3. **Invoke** — the exact next invocation and minimum input.
4. **Freshness check** — only when a product capability may have changed.

Use `/knowledge-pack:<skill>` for the Claude Code plugin and `/<skill>` for a
standalone Claude skill. Use `$knowledge-pack:<skill>` for the Codex plugin and
`$<skill>` for a standalone Codex skill.

## Boundaries

- Choose one adequate route, not a menu of equivalent options.
- Own comparisons such as “custom agent or skill?”; once the user explicitly selects an artifact, its specialist skill wins.
- Do not execute the specialist workflow unless routing alone would block progress.
- Do not conflate a skill, subagent, hook, plugin, connector, or MCP server.
- Answer in the user's language.

Files in this skill

  • SKILL.md2.3 KB
  • agents/openai.yaml239 B
  • references/knowledge-base.md3.6 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.