# State And Attribute Change Tracking

> Use this skill when the user asks for questions about something changing over time, such as melting, darkening, opening, shrinking, turning on, or switching state. Trigger it for requests like 'make the clip look the same except for one change', 'ask what changed', 'did it open or close', or 'make the answer depend on the before-versus-after state.'

- Skill: `dingxingdi/state-and-attribute-change-tracking-2` (Agent Skill, multi-file: 5 files)
- Install (CLI): `npx skillmds@latest add dingxingdi/state-and-attribute-change-tracking-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/dingxingdi/state-and-attribute-change-tracking-2/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: dingxingdi (https://skillmd.com/u/dingxingdi)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/dingxingdi/state-and-attribute-change-tracking-2

---


# Skill: state-and-attribute-change-tracking
## 1. Capability Definition & Real Case
* **Professional Definition**: The capability to detect how an entity or the scene changes over time, including color, light, size, shape, material state, or on/off status, and to distinguish real temporal change from static appearance.
* **Dimension Hierarchy**: Temporal-Spatial Understanding->Dynamic Event Perception->state-and-attribute-change-tracking
### Real Case

**[Case 1]**
* **Initial Environment**: A 10-second close-up video of a vanilla soft-serve cone on a table near direct sunlight. The framing stays almost fixed, and there is no audio cue that reveals the answer.
* **Real Question**: What is happening to the ice cream?
* **Real Trajectory**: Compare the contour of the top swirl across early, middle, and late frames, inspect the gradual loss of edge sharpness, and verify that the liquid runoff increases rather than the object becoming more rigid or opaque.
* **Real Answer**: Melting.
* **Why this demonstrates the capability**: A single frame could still look like ordinary ice cream, so the answer depends on observing progressive state change across time. The capability being tested is the model's ability to track attribute evolution instead of over-relying on static visual priors.

---
## 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:

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`