Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Navigate To

ASecurity

Drives a vehicle to a destination autonomously using CARLA's navigation agents (BasicAgent/BehaviorAgent/ConstantVelocityAgent), or plans a route between two points with GlobalRoutePlanner. The agent handles routing, obstacle avoidance and traffic-light obedience. Use when the user asks to "drive the ego to (x,y,z)", "navigate to the junction", "go to that location", "plan a route from A to B", or "make the car drive itself somewhere specific". Manual control is control-vehicle; autopilot roa...

3 stars
0 votes
0 copies
0 views
Added 9/20/2026
ai-agentspythongobashapi

Works with

api

Security Analysis

A100/100

Scanned 9/20/2026

Install to Claude Code

$npx -y skills add carla-simulator/carla-agentic-tools --skill navigate-to --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Navigate To?

Add the live security badge to your README — it updates automatically with every re-scan.

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

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: navigate-to
description: Drives a vehicle to a destination autonomously using CARLA's navigation agents (BasicAgent/BehaviorAgent/ConstantVelocityAgent), or plans a route between two points with GlobalRoutePlanner. The agent handles routing, obstacle avoidance and traffic-light obedience. Use when the user asks to "drive the ego to (x,y,z)", "navigate to the junction", "go to that location", "plan a route from A to B", or "make the car drive itself somewhere specific". Manual control is control-vehicle; autopilot roaming is control-traffic.
license: MIT
compatibility: Any OS with the CARLA PythonAPI installed and the CARLA checkout's PythonAPI/carla/agents on PYTHONPATH (env.sh sets it from CARLA_ROOT), plus a reachable running server with a target vehicle. Does NOT need UE4_ROOT. Tested against CARLA 0.9.16.
metadata:
  group: python-api
  prerequisites: scripts/check_env.sh
  reference: references/navigation.md
---

# Navigate a vehicle to a destination

> **Paths.** `scripts/…` and `references/…` below are relative to the
> directory holding this SKILL.md. Your working directory is the user's
> project, not that directory, so prefix them with its absolute path or the
> command is not found.

Point-to-point autonomous driving. `go` drives the ego to a destination with a
navigation agent that plans a route and follows it while avoiding obstacles and
obeying (or ignoring, on request) traffic lights. `route` just plans the path.

This is distinct from the sibling driving skills: [`control-vehicle`](../control-vehicle/SKILL.md)
is manual actuator input; [`control-traffic`](../control-traffic/SKILL.md) is the
Traffic Manager making cars roam; this drives one car to a **specific place**.

## Instructions

```
Progress:
- [ ] Step 1: Check prerequisites (bash scripts/check_env.sh), clear FAILs (incl. agents import)
- [ ] Step 2: Have an ego (spawn-vehicles ego) and a destination (world-data/map-waypoints)
- [ ] Step 3: go (drive) — or route (plan only). Runs a tick loop until arrival/timeout
```

Commands need `CARLA_HOST`/`CARLA_PORT` + `CARLA_ROOT` from `scripts/env.sh`.
**Negative coordinates:** use the `=` form (`--to=-24,-57,0.6`).

### Step 1: Check prerequisites

```bash
bash scripts/check_env.sh
```

FAILs if the `agents` package isn't importable — set `CARLA_ROOT` to a carla
checkout so `PythonAPI/carla/agents` is on `PYTHONPATH`.

### Step 3: Navigate

```bash
source scripts/env.sh

# plan a route and draw it (no driving)
python3 scripts/navigate.py route --from=0,0,1 --to=-24,-57,1 --draw

# drive the ego there (BasicAgent, obeys lights)
python3 scripts/navigate.py go --to=-24,-57,1 --speed 25

# a cautious/aggressive style
python3 scripts/navigate.py go --to=-24,-57,1 --agent behavior --behavior aggressive

# ignore lights (e.g. an emergency run) and give it more time
python3 scripts/navigate.py go --to=100,20,1 --ignore-lights --seconds 120
```

`go` targets the ego (`role hero`) by default; `--id`/`--filter` for another
vehicle. It blocks until the car arrives (`agent.done()`) or `--seconds` elapses.

## Examples

**Example 1: drive to a spot**

User says: "drive the ego to (−24, −57)"

`go --to=-24,-57,1`. It plans, drives, and prints "arrived" on success.

**Example 2: go to a resolved element**

User says: "drive to the 4-way junction in the middle"

`map-waypoints junctions --arms 4` → pick its centre → `go --to=<centre>`.

**Example 3: just the route**

User says: "what's the route from here to the depot?"

`route --from=<here> --to=<depot> --draw` → maneuver list + drawn path.

## Troubleshooting

**Problem: `cannot import agents`**
Cause: the agents package (checkout-only) isn't on PYTHONPATH.
Solution: set `CARLA_ROOT` to a carla checkout; env.sh adds `PythonAPI/carla`.

**Problem: it never arrives / stops at timeout**
Cause: unreachable destination, too little time, or stuck in traffic.
Solution: verify the destination is on a road (map-waypoints `waypoint`), raise
`--seconds`, or `--ignore-vehicles` to push through.

**Problem: the car doesn't move in sync mode**
Cause: `go` ticks the world itself in sync mode; if another process also ticks,
they fight.
Solution: run one ticker. In async it just drives.

**Problem: it stops at every red light forever**
Cause: dense traffic / lights.
Solution: `--ignore-lights` (and/or the control-traffic-lights skill to free them).

## Outputs

Server state — the vehicle drives to the destination (or a planned route + drawn
overlay for `route`). No file. `go` reports arrival or timeout.

Detail (agent types, route planner, run-step loop, sync interaction) in
[references/navigation.md](references/navigation.md).

Attribution

carla-simulatorcarla-simulator
View sourceMore from carla-simulator →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

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 (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Caveman

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

1023331 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3331 votes

catchup

Recovers prior coding-agent session context by running `catchup <agent> --since-compact`, which extracts a clean summary of a previous Codex, Claude Code, Antigravity, OpenCode, or Pi Agent session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", or asks to recover/summarize a previous session before continuing. Do NOT use for the current conversation, git history, or any non-agent log.

611 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →