How do I install the Symbolic State Tracking skill?

Run npx skillmds@latest add dingxingdi/symbolic-state-tracking-2 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Symbolic State Tracking safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Symbolic State Tracking?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Symbolic State Tracking free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Symbolic State Tracking?

dingxingdi (@dingxingdi) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Symbolic State Tracking by dingxingdi

Name: Symbolic State Tracking
Author: dingxingdi

Symbolic State Tracking

Use this when the user wants planning data about keeping track of where things are, what is on top of what, what is free or blocked, or when later steps only work if earlier moves changed the state correctly. Trigger it for requests like 'make tasks where one wrong early move ruins everything', 'give me block moving plans that depend on what is clear', 'make planning problems about tracking changing object states', or 'create tasks where the plan must remember what changed after each step.'

dingxingdi Updated Sep 17, 2026

File contents

Skill: symbolic-state-tracking

1. Capability Definition & Real Case

Professional Definition: The ability to maintain an internally consistent symbolic model of object relations, action preconditions, and action effects across a multi-step plan so that every subsequent action is grounded in the state produced by earlier actions.
Dimension Hierarchy: Closed-World Symbolic Planning->State and Resource Reasoning->symbolic-state-tracking

Real Case

[Case 1]

Initial Environment: A symbolic tabletop world contains five blocks. The planner is told that b5 is on b3, b4 is on b2, b2 is on b1, b3 is on b4, b1 is on the table, b5 is clear, and the robot arm is empty.
Real Question: You have 5 blocks. One cannot place more than one block on another block. b5 is on top of b3. b4 is on top of b2. b2 is on top of b1. b3 is on top of b4. b1 is on the table. b5 is clear. Your arm is empty. Your goal is to move the blocks. b1 should be on top of b2. b3 should be on top of b5. b4 should be on top of b1.
Real Trajectory: Unstack b5 from b3 and put it on the table. Unstack b3 from b4 and stack it on b5. Unstack b4 from b2 and put it on the table. Unstack b2 from b1 and put it on the table. Pick up b1 and stack it on b2. Pick up b4 and stack it on b1.
Real Answer: A valid final plan is: unstack(b5,b3) -> putdown(b5) -> unstack(b3,b4) -> stack(b3,b5) -> unstack(b4,b2) -> putdown(b4) -> unstack(b2,b1) -> putdown(b2) -> pickup(b1) -> stack(b1,b2) -> pickup(b4) -> stack(b4,b1).
Why this demonstrates the capability: The task cannot be solved by local pattern matching because the planner must update the truth of ON, CLEAR, ARM-EMPTY, and holding predicates after every move. Each later action depends on whether prior actions created the required clear surfaces and free gripper state. This makes the benchmark sensitive to state-update errors rather than mere goal recognition.

Pipeline Execution Instructions

To synthesize data for this capability, you must strictly follow a 3-phase pipeline. Do not hallucinate steps. Read the corresponding reference file for each phase sequentially:

Phase 1: Environment Exploration Read the exploration guidelines to discover raw knowledge seeds: references/EXPLORATION.md
Phase 2: Trajectory Selection Once Phase 1 is complete, read the selection criteria to evaluate the trajectory: references/SELECTION.md
Phase 3: Data Synthesis Once a trajectory passes Phase 2, read the synthesis instructions to generate the final data: references/SYNTHESIS.md

---name: symbolic-state-tracking-2description: Use this when the user wants planning data about keeping track of where things are, what is on top of what, what is free or blocked, or when later steps only work if earlier moves changed the state correctly. Trigger it for requests like 'make tasks where one wrong early move ruins everything', 'give me block moving plans that depend on what is clear', 'make planning problems about tracking changing object states', or 'create tasks where the plan must remember what changed after each step.'---# Skill: symbolic-state-tracking## 1. Capability Definition & Real Case* **Professional Definition**: The ability to maintain an internally consistent symbolic model of object relations, action preconditions, and action effects across a multi-step plan so that every subsequent action is grounded in the state produced by earlier actions.* **Dimension Hierarchy**: Closed-World Symbolic Planning->State and Resource Reasoning->symbolic-state-tracking### Real Case**[Case 1]*** **Initial Environment**: A symbolic tabletop world contains five blocks. The planner is told that b5 is on b3, b4 is on b2, b2 is on b1, b3 is on b4, b1 is on the table, b5 is clear, and the robot arm is empty.* **Real Question**: You have 5 blocks. One cannot place more than one block on another block. b5 is on top of b3. b4 is on top of b2. b2 is on top of b1. b3 is on top of b4. b1 is on the table. b5 is clear. Your arm is empty. Your goal is to move the blocks. b1 should be on top of b2. b3 should be on top of b5. b4 should be on top of b1.* **Real Trajectory**: Unstack b5 from b3 and put it on the table. Unstack b3 from b4 and stack it on b5. Unstack b4 from b2 and put it on the table. Unstack b2 from b1 and put it on the table. Pick up b1 and stack it on b2. Pick up b4 and stack it on b1.* **Real Answer**: A valid final plan is: unstack(b5,b3) -> putdown(b5) -> unstack(b3,b4) -> stack(b3,b5) -> unstack(b4,b2) -> putdown(b4) -> unstack(b2,b1) -> putdown(b2) -> pickup(b1) -> stack(b1,b2) -> pickup(b4) -> stack(b4,b1).* **Why this demonstrates the capability**: The task cannot be solved by local pattern matching because the planner must update the truth of ON, CLEAR, ARM-EMPTY, and holding predicates after every move. Each later action depends on whether prior actions created the required clear surfaces and free gripper state. This makes the benchmark sensitive to state-update errors rather than mere goal recognition.## Pipeline Execution InstructionsTo synthesize data for this capability, you must strictly follow a 3-phase pipeline. **Do not hallucinate steps.** Read the corresponding reference file for each phase sequentially:1. **Phase 1: Environment Exploration** Read the exploration guidelines to discover raw knowledge seeds: `references/EXPLORATION.md`2. **Phase 2: Trajectory Selection** Once Phase 1 is complete, read the selection criteria to evaluate the trajectory: `references/SELECTION.md`3. **Phase 3: Data Synthesis** Once a trajectory passes Phase 2, read the synthesis instructions to generate the final data: `references/SYNTHESIS.md`

Frequently asked questions