Trackpy Particle Tracking
Single-particle tracking in microscopy video: locate, link, filter, MSD diffusion with trackpy
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 31, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 3a575ed
Fetched SKILL.md (only file in the dir; no bundled scripts, no security smells; frontmatter has name+description, license BSD-3). Installed trackpy 0.7 + pims in a venv, generated a synthetic 60-frame TIF of 25 Gaussian particles doing a random walk with known D_true=0.8 px^2/frame. Ran a BASELINE script (guessed diameter, default minmass, no stub filter, no drift correction, crude lag-1 MSD) and a SKILL script (Workflow 1 verbatim: locate tune, batch processes=1, link+memory, filter_stubs, compute_drift/subtract_drift, emsd linear-regime fit). Baseline over-detected noise (~55 features/frame vs 25 real), fragmented into 1752 tracks, and returned D=1.259 (57% high); the skill filtered to 20 clean tracks and returned D=0.691 (R^2=0.998), much closer to truth. All documented API calls (locate, batch, link, filter_stubs, compute_drift, subtract_drift, emsd) executed correctly.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 8/10
- Docs & honesty 5/5
What Trackpy Particle Tracking does
A reference skill for the Python trackpy library that walks Claude through single-particle tracking (SPT) in video microscopy: locating bright spots per frame, linking them into trajectories, filtering short tracks, correcting stage drift, and computing MSD to extract diffusion coefficients in 2D and 3D. Triggers when a user has a fluorescence or brightfield microscopy video (TIF stack, AVI, image series) and wants per-particle trajectories, mean squared displacement curves, or diffusion coefficients. Does not trigger for image segmentation, cell counting, or generic object detection.
How to install Trackpy Particle Tracking
git clone --depth 1 https://github.com/jaechang-hits/SciAgent-Skills.git /tmp/trackpy-particle-tracking-src
mkdir -p ~/.claude/skills
cp -R /tmp/trackpy-particle-tracking-src/skills/cell-biology/trackpy-particle-tracking ~/.claude/skills/trackpy-particle-tracking
# Runtime deps (skill is doc-only, no bundled scripts):
# pip install trackpy pims pandas numpy matplotlib scipy
# For ND2/CZI/LIF microscopy formats: pip install pims-nd2 OR pip install aicsimageio
# Verified with trackpy 0.7 in a clean venv.
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Trackpy Particle Tracking
-
/trackpy-particle-trackingSingle-particle tracking in microscopy video: locate, link, filter, MSD diffusion with trackpy
It also activates on plain-language prompts like these:
-
Track fluorescent particles in this microscopy video -
Compute MSD for these vesicle trajectories -
Link detected spots into particle tracks across frames
Frequently asked questions
- Is the Trackpy Particle Tracking skill free?
- Yes. The skill itself is free from jaechang-hits/SciAgent-Skills. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Trackpy Particle Tracking work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 31, 2026. Verdict: Tested · Works. Fetched SKILL.md (only file in the dir; no bundled scripts, no security smells; frontmatter has name+description, license BSD-3). Installed trackpy 0.7 + pims in a venv, generated a synthetic 60-frame TIF of 25 Gaussian particles doing a random walk with known D_true=0.8 px^2/frame. Ran a BASELINE script (guessed diameter, default minmass, no stub filter, no drift correction, crude lag-1 MSD) and a SKILL script (Workflow 1 verbatim: locate tune, batch processes=1, link+memory, filter_stubs, compute_drift/subtract_drift, emsd linear-regime fit). Baseline over-detected noise (~55 features/frame vs 25 real), fragmented into 1752 tracks, and returned D=1.259 (57% high); the skill filtered to 20 clean tracks and returned D=0.691 (R^2=0.998), much closer to truth. All documented API calls (locate, batch, link, filter_stubs, compute_drift, subtract_drift, emsd) executed correctly.
- What is the Trackpy Particle Tracking SkillProof Score?
- 9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.
- How do I install Trackpy Particle Tracking?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Trackpy Particle Tracking with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.