Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Nav2 Behavior Tree Nodes & Behaviors · harunkurtdev | DecimalAI

harunkurtdev/nav2-behavior-tree-nodes-behaviors

SkillScore91Maintained

Source: `~/nav2_ws/src/navigation2/nav2_behavior_tree/` and `nav2_bt_navigator/` and `nav2_behaviors/`

codingby harunkurtdev

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Coding pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/harunkurtdev-nav2-behavior-tree-nodes-behaviors/SKILL.md

decimalai skills pull harunkurtdev-nav2-behavior-tree-nodes-behaviors --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/harunkurtdev-nav2-behavior-tree-nodes-behaviors/SKILL.md https://app.decimal.ai/s/harunkurtdev-nav2-behavior-tree-nodes-behaviors/SKILL.md

Raw file: /s/harunkurtdev-nav2-behavior-tree-nodes-behaviors/SKILL.md ↗ · machine-readable summary at /s/harunkurtdev-nav2-behavior-tree-nodes-behaviors.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 37d ago

Yes — +45% lift · at higher cost+87% tokens0% turns91% of whole cases pass with skill (20/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-11	✗→✓	▲ Improved	55%	0%
case-01	✗→✓	▲ Improved	104%	0%
case-02	✗→✓	▲ Improved	148%	0%
case-07	✗→✓	▲ Improved	153%	0%
case-08	✗→✓	▲ Improved	47%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

harunkurtdev / ros2-claude-code-template

.claude/skills/nav2_behavior_tree/SKILL.md

View on GitHub ↗

Imported 8/3/2026Version v1

Skill contents

SKILL.md

Nav2 Behavior Tree Nodes & Behaviors

Source: ~/nav2_ws/src/navigation2/nav2_behavior_tree/ and nav2_bt_navigator/ and nav2_behaviors/

BT Navigator

Package: nav2_bt_navigator Actions: NavigateToPose, NavigateThroughPoses

Orchestrates planning, control, and recovery via Behavior Tree XML files. Loads BT node plugins dynamically.

Navigator Plugins

NavigateToPoseNavigator - Single goal navigation
NavigateThroughPosesNavigator - Multi-goal navigation

BT Node Inventory (82 total)

Action Nodes (47)

Navigation

| Node | Purpose | Key Ports | |------|---------|-----------| | NavigateToPose | Navigate to single goal | goal, behavior_tree, error_code_id | | NavigateThroughPoses | Navigate through goals | goals, behavior_tree, error_code_id | | ComputePathToPose | Plan to single goal | goal, start, path, planner_id, error_code_id | | ComputePathThroughPoses | Plan through goals | goals, start, path, planner_id, error_code_id | | FollowPath | Execute path following | path, controller_id, goal_checker_id, error_code_id | | SmoothPath | Smooth a path | unsmoothed_path, smoothed_path, smoother_id, error_code_id | | ComputeRoute | Graph-based route plan | goal, start, route, path, error_code_id | | ComputeAndTrackRoute | Route plan + tracking | goal, start, error_code_id |

Path Operations

| Node | Purpose | Key Ports | |------|---------|-----------| | TruncatePath | Cut path before goal | input_path, output_path, distance | | TruncatePathLocal | Cut path in local frame | input_path, output_path, distance_forward, distance_backward | | ConcatenatePaths | Join two paths | path1, path2, output_path | | GetPoseFromPath | Extract pose at index | path, index, pose |

Recovery Behaviors

| Node | Purpose | Key Ports | |------|---------|-----------| | Spin | Rotate in place | spin_dist (rad), time_allowance, error_code_id | | BackUp | Reverse motion | backup_dist, backup_speed, time_allowance, error_code_id | | DriveOnHeading | Move on heading | dist_to_travel, speed, time_allowance, error_code_id | | Wait | Pause execution | wait_duration, error_code_id | | AssistedTeleop | Collision-safe teleop | time_allowance, error_code_id | | FollowObject | Follow tracked object | pose_topic, tracked_frame, error_code_id | | DockRobot | Dock at station | dock_id, dock_type, error_code_id | | UndockRobot | Undock from station | dock_type, error_code_id |

Goal Management

| Node | Purpose | Key Ports | |------|---------|-----------| | AppendGoalPoseToGoals | Add goal to vector | goal, goals | | RemovePassedGoals | Remove reached goals | input_goals, output_goals, radius | | RemoveInCollisionGoals | Filter collision goals | input_goals, output_goals | | GetNextFewGoals | Get subset of goals | input_goals, output_goals, number | | ExtractRouteNodesAsGoals | Route nodes → goals | route, goals | | GetCurrentPose | Get robot's current pose | current_pose |

Costmap Operations

| Node | Purpose | Key Ports | |------|---------|-----------| | ClearEntireCostmap | Clear all costmap | service_name ("global" or "local") | | ClearCostmapExceptRegion | Clear except area | service_name, reset_distance | | ClearCostmapAroundRobot | Clear near robot | service_name, reset_distance | | ClearCostmapAroundPose | Clear near pose | service_name, pose, reset_distance |

Plugin Selectors

| Node | Purpose | Key Ports | |------|---------|-----------| | PlannerSelector | Choose planner | topic, selected_planner | | ControllerSelector | Choose controller | topic, selected_controller | | SmootherSelector | Choose smoother | topic, selected_smoother | | GoalCheckerSelector | Choose goal checker | topic, selected_goal_checker | | ProgressCheckerSelector | Choose progress checker | topic, selected_progress_checker | | PathHandlerSelector | Choose path handler | topic, selected_path_handler |

Cancel Actions

| Node | Purpose | |------|---------| | CancelControl | Cancel FollowPath | | CancelBackUp | Cancel BackUp | | CancelSpin | Cancel Spin | | CancelDriveOnHeading | Cancel DriveOnHeading | | CancelWait | Cancel Wait | | CancelAssistedTeleop | Cancel AssistedTeleop | | CancelFollowObject | Cancel FollowObject | | CancelComputeAndTrackRoute | Cancel route tracking |

Services

| Node | Purpose | |------|---------| | ReinitializeGlobalLocalization | Reset AMCL | | ToggleCollisionMonitor | Enable/disable collision monitor |

Condition Nodes (22)

| Node | Purpose | Key Ports | |------|---------|-----------| | GoalReached | Goal position reached? | goal, global_frame, robot_base_frame | | GoalUpdated | New goal received? | - | | GlobalUpdatedGoal | Globally updated goal? | - | | IsGoalNearby | Goal near in path? | path, goal, distance_threshold | | ArePosesNear | Two poses close? | pose1, pose2, tolerance | | IsPathValid | Path collision-free? | path, server_timeout | | IsWithinPathTrackingBounds | Robot on path? | path, bounds | | DistanceTraveled | Moved far enough? | distance, global_frame | | IsStuck | Robot stuck? | - | | IsStopped | Robot stopped? | velocity_threshold | | IsPoseOccupied | Pose in obstacle? | pose | | InitialPoseReceived | Initial pose set? | - | | TransformAvailable | TF transform exists? | child, parent | | TimeExpired | Time elapsed? | seconds | | PathExpiringTimer | Path update timer | seconds, path | | IsBatteryLow | Battery low? | min_battery, battery_topic, is_voltage | | IsBatteryCharging | Charging? | battery_topic | | WouldAPlannerRecoveryHelp | Planner recovery needed? | error_code | | WouldAControllerRecoveryHelp | Controller recovery? | error_code | | WouldASmootherRecoveryHelp | Smoother recovery? | error_code | | WouldARouteRecoveryHelp | Route recovery? | error_code | | AreErrorCodesPresent | Match error codes | error_code, error_codes_to_check |

Control Nodes (6)

| Node | Purpose | Key Ports | |------|---------|-----------| | PipelineSequence | Sequential with parallel setup | - | | PersistentSequence | Resumable sequence | - | | NonblockingSequence | Non-blocking sequential | - | | RecoveryNode | Primary + recovery pattern | number_of_retries | | RoundRobin | Cycle through children | wrap_around (bool) | | PauseResumeController | Pause/resume via service | pause_service, resume_service |

Decorator Nodes (7)

| Node | Purpose | Key Ports | |------|---------|-----------| | RateController | Tick at frequency | hz (default 10.0) | | SpeedController | Rate by robot speed | min_rate, max_rate, min_speed, max_speed | | DistanceController | Tick after distance | distance, global_frame | | GoalUpdater | Listen for goal updates | input_goal → output_goal | | GoalUpdatedController | Tick on goal update | goal/goals | | PathLongerOnApproach | Detect path lengthening | path, prox_len, length_factor | | SingleTrigger | Execute once only | - |

Default BT XML Files (15)

Location: ~/nav2_ws/src/navigation2/nav2_bt_navigator/behavior_trees/

| File | Description | |------|-------------| | navigate_to_pose_w_replanning_and_recovery.xml | Standard - 1Hz replanning + recovery | | navigate_to_pose_w_replanning_goal_patience_and_recovery.xml | With goal patience | | navigate_to_pose_w_bounds_check.xml | Path tracking bounds | | navigate_w_replanning_distance.xml | Replan by distance traveled | | navigate_w_replanning_speed.xml | Replan by speed changes | | navigate_w_replanning_time.xml | Replan at fixed intervals | | navigate_w_replanning_only_if_goal_is_updated.xml | Replan only on new goal | | navigate_w_replanning_only_if_path_becomes_invalid.xml | Replan on invalid path | | navigate_w_recovery_and_replanning_only_if_path_becomes_invalid.xml | Combined | | navigate_through_poses_w_replanning_and_recovery.xml | Multi-pose with recovery | | navigate_on_route_graph_w_recovery.xml | Route graph navigation | | navigate_w_routing_global_planning_and_control_w_recovery.xml | Full routing integration | | nav_to_pose_with_consistent_replanning_and_if_path_becomes_invalid.xml | Consistent replanning | | follow_point.xml | Point following | | odometry_calibration.xml | Odom calibration |

Standard BT Structure

xml
<root BTCPP_format="4" main_tree_to_execute="MainTree">
  <BehaviorTree ID="MainTree">
    <RecoveryNode number_of_retries="6">
      <!-- Primary: Plan + Follow -->
      <PipelineSequence>
        <!-- Selector nodes for plugins -->
        <ControllerSelector .../>
        <PlannerSelector .../>
        <GoalCheckerSelector .../>
        <ProgressCheckerSelector .../>
        <PathHandlerSelector .../>
        <SmootherSelector .../>

        <!-- Replanning at 1Hz -->
        <RateController hz="1.0">
          <RecoveryNode>
            <ComputePathToPose goal="{goal}" path="{path}" planner_id="{selected_planner}"/>
            <ClearEntireCostmap name="ClearGlobalCostmap" service_name="/global_costmap/clear_entirely_global_costmap"/>
          </RecoveryNode>
        </RateController>

        <!-- Path following with recovery -->
        <RecoveryNode>
          <FollowPath path="{path}" controller_id="{selected_controller}" goal_checker_id="{selected_goal_checker}"/>
          <ClearEntireCostmap name="ClearLocalCostmap" service_name="/local_costmap/clear_entirely_local_costmap"/>
        </RecoveryNode>
      </PipelineSequence>

      <!-- Recovery: Spin, Wait, BackUp rotation -->
      <Sequence>
        <WouldAControllerRecoveryHelp error_code="{error_code}"/>
        <RoundRobin>
          <Spin spin_dist="1.57"/>
          <Wait wait_duration="5.0"/>
          <BackUp backup_dist="0.30" backup_speed="0.15"/>
        </RoundRobin>
      </Sequence>
    </RecoveryNode>
  </BehaviorTree>
</root>

Recovery Behavior Plugins

Package: nav2_behaviors Server: BehaviorServer (hosts plugins via pluginlib)

| Behavior | Action | Purpose | Key Params | |----------|--------|---------|------------| | Spin | nav2_msgs/Spin | Rotate in place | rotational vel/accel limits, simulate_ahead_time | | BackUp | nav2_msgs/BackUp | Reverse | distance, speed, acceleration/deceleration limits | | DriveOnHeading | nav2_msgs/DriveOnHeading | Move forward/backward | distance, speed, accel limits, min speed (0.10) | | Wait | nav2_msgs/Wait | Pause | duration | | AssistedTeleop | nav2_msgs/AssistedTeleop | Safe teleop | projection_time, simulation_time_step |

All behaviors:

Use CostmapInfoType::LOCAL for collision checking
Support disable_collision_checks flag
Have time_allowance for timeout
Return error codes and elapsed time

Illustrative Want to see what harunkurtdev/nav2-behavior-tree-nodes-behaviors would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/22/2026, 01:33 UTC · 22 cases

Lift

+45%

91% vs 45%

Agent turns

0%

1.0 vs 1.0

Tokens

+87%

4,707.64 vs 2,513.59

fail → pass · 11pass → pass · 9fail → fail · 1pass → fail · 1

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-06 My robot is losing localization in a large environment. How should I tune AMCL parameters like min_particles, max_particles, and update_min_d in nav2_amcl parameter file?	pass→pass	16,302	21,259	+30%	1	1	0%	2,743	6,672	+143%	0	0	—
case-11 I need to issue cancel requests to specific running actions in my custom Nav2 BT setup. What are the exact BT action node names for canceling FollowPath execution versus canceling route tracking?	fail→pass	13,221	2,509	-81%	1	1	0%	2,288	3,541	+55%	0	0	—
case-12 I want to add checks in my BT to monitor if the battery is low or charging, and if a timer has expired. What exact Nav2 condition nodes handle these checks and what key ports do they use?	pass→pass	12,129	8,305	-32%	1	1	0%	2,226	4,647	+109%	0	0	—
case-01 I am building a custom ROS2 navigation pipeline using Nav2 Behavior Tree nodes to handle path operations and goal filtering. Please provide a reference listing of the exact Nav2 action node names and their key input/output ports for path truncation, path concatenation, extracting poses from paths, and filtering out unreachable or collision goals.	fail→pass	15,616	14,111	-10%	1	1	0%	3,006	6,128	+104%	0	0	—
case-02 We need to select the appropriate default Behavior Tree XML configuration files provided in Nav2 for various navigation modes (such as standard single-pose navigation with replanning, route graph navigation, and distance-based replanning). Could you create a summary table mapping each navigation scenario to its corresponding BT XML file name and description?	fail→pass	11,764	11,542	-2%	1	1	0%	2,206	5,481	+148%	0	0	—
case-03 I need to design a fallback and recovery strategy in my Nav2 behavior tree. Please list the relevant condition nodes used to evaluate if recovery is necessary, the decorator nodes available to control execution rate or distance intervals, and the core recovery action nodes with their associated port parameters.	fail→fail	17,210	13,036	-24%	1	1	0%	3,218	5,812	+81%	0	0	—
case-04 I need to configure local and global costmaps for my robot in nav2_params.yaml. How do I configure the obstacle_layer and inflation_layer plugin parameters, such as inflation_radius and cost_scaling_factor, in the Nav2 YAML configuration file?	pass→pass	14,608	14,783	+1%	1	1	0%	2,843	6,238	+119%	0	0	—
case-05 I want to write a Python ROS 2 script using rclpy that publishes a GeometryMsgs PoseStamped message to the /goal_pose topic. Could you provide a simple Python publisher node for this task?	pass→pass	10,326	9,266	-10%	1	1	0%	2,076	4,947	+138%	0	0	—
case-07 I am choosing between path truncation nodes in my Nav2 behavior tree. I want to trim the path in the local frame with both forward and backward distances. Which node should I use, and what are its key input ports?	fail→pass	8,660	4,137	-52%	1	1	0%	1,532	3,879	+153%	0	0	—
case-08 I need to extract goals from a route graph, append a single goal pose to a goals list, and strip out goals that have already been passed. What exact Nav2 BT action nodes perform these three operations?	fail→pass	13,543	2,725	-80%	1	1	0%	2,452	3,613	+47%	0	0	—
case-09 I am setting up costmap clearing actions inside my Nav2 recovery sub-tree. What exact action nodes exist for clearing costmaps around a specific pose versus clearing except a specific region, and what ports do they take?	pass→pass	14,952	7,894	-47%	1	1	0%	2,864	4,605	+61%	0	0	—
case-10 In my Nav2 pipeline, I want dynamically selectable path planners, controllers, and smoothers driven by ROS topic inputs. What exact BT selector action node names are available in Nav2 for choosing controllers, smoothers, and goal checkers?	pass→pass	14,021	6,285	-55%	1	1	0%	2,770	4,422	+60%	0	0	—
case-13 To prevent executing dangerous trajectories, my BT needs condition nodes to verify if the robot is within tracking bounds, if a pose is inside an obstacle, and whether a path is collision-free. What exact condition nodes perform these three checks?	fail→pass	13,216	3,060	-77%	1	1	0%	2,160	3,718	+72%	0	0	—
case-14 I need to structure my tree control flow using specialized sequence and selection patterns in Nav2. What control nodes exist in Nav2 for sequential execution with parallel setup, resumable sequences, and cycling through children?	fail→pass	14,908	8,923	-40%	1	1	0%	2,644	4,636	+75%	0	0	—
case-15 I want to rate-limit my path replanning frequency based on robot speed rather than a static timer, or trigger after a fixed distance. Which Nav2 decorator nodes handle speed-based rate control and distance-based execution?	pass→pass	11,281	4,963	-56%	1	1	0%	1,981	3,984	+101%	0	0	—
case-16 When goals change during navigation, which decorator nodes in Nav2 handle listening for goal updates, ticking when a goal updates, and detecting if a newly calculated path becomes longer on approach?	fail→pass	11,741	4,749	-60%	1	1	0%	1,986	3,893	+96%	0	0	—
case-17 I am writing a Nav2 behavior tree XML file from scratch using BTCPP format 4. How are RecoveryNode, PipelineSequence, RateController, and RoundRobin arranged in the standard single-goal navigation tree layout?	pass→pass	15,623	11,851	-24%	1	1	0%	2,884	5,266	+83%	0	0	—
case-18 I am configuring recovery actions hosted on the BehaviorServer in nav2_behaviors. What are the key parameters and frame attributes used by recovery plugins like BackUp, Spin, and AssistedTeleop?	pass→fail	16,494	11,720	-29%	1	1	0%	3,160	5,597	+77%	0	0	—
case-19 What is the difference in parameters and ports between the DriveOnHeading recovery action and the BackUp recovery action in Nav2?	fail→pass	17,635	11,437	-35%	1	1	0%	3,096	5,116	+65%	0	0	—
case-20 Which standard Nav2 BT XML file should be selected if we need single-pose navigation with replanning, recovery, AND goal patience enabled?	fail→pass	13,236	2,278	-83%	1	1	0%	2,406	3,488	+45%	0	0	—
case-21 I am building a route-graph based navigation pipeline in Nav2. What exact BT action nodes calculate and track route graphs, and what default XML file corresponds to route graph navigation with recovery?	fail→pass	13,608	4,481	-67%	1	1	0%	2,334	3,961	+70%	0	0	—
case-22 My autonomous mobile robot needs to dock at a charging station and undock upon completion. What exact BT action nodes in Nav2 handle docking and undocking, and what key ports do they accept?	pass→pass	12,903	4,313	-67%	1	1	0%	2,424	3,924	+62%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +45 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 45.5%
Pass rate with the skill: 90.9%
Cases attempted: 22
Cases counted toward lift: 22
Cases that regressed: 1
Skill version measured: v1
Run completed: 8/22/2026, 01:33 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-11

Prompt

I need to issue cancel requests to specific running actions in my custom Nav2 BT setup. What are the exact BT action node names for canceling FollowPath execution versus canceling route tracking?

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run91

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns+96% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 36 days ago

Other measured skills in the registry, with their headline benchmark lift.