Agent skill

Sleap Support

by talmolab in talmolab/sleap

Handle SLEAP GitHub support workflow for issues and discussions.

BSD-3-Clause-ClearAuto-check passedDevelopment

Install Sleap Support

skills CLI
$ npx skills add talmolab/sleap --skill sleap-support -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install talmolab/sleap sleap-support --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/talmolab/sleap.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/sleap-support .claude/skills/sleap-support && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
sleap-support
GitHub stars
618
Token cost
~2k tokens
SKILL.md length
460 words
Files
4
Skills in repo
4
Repo updated
First seen
Licence
BSD-3-Clause-Clear

At a glance

Handle SLEAP GitHub support workflow for issues and discussions.

  • Works in 9 steps: Create Investigation Folder → Fetch Post Content → Download Images → …
  • The user says support
  • SKILL.md covers Quick Start, Workflow Steps, Response Tone Guidelines and When Requesting User Actions, plus 3 more sections
  • Calls gh, git and wget; reaches slp.sh

What it does

Sleap Support is an agent skill from talmolab/sleap. Handle SLEAP GitHub support workflow for issues and discussions. Use when the user says "support", provides a GitHub issue/discussion number like "2512", or asks to investigate a user report from talmolab/sleap. Scaffolds investigation folders, downloads posts with images, analyzes problems, and drafts friendly responses.

Its SKILL.md is about 2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files (for example `gh-commands.md`, `response-templates.md` and `troubleshooting-patterns.md`).

It sits in Development, covering Project scaffolding. It works with GitHub. The repository describes itself as: A deep learning framework for multi-animal pose tracking.

When your agent uses it

  • The user says support
  • Provides a GitHub issue/discussion number like 2512
  • Asks to investigate a user report from talmolab/sleap

Example prompts

  • “support”
  • “/sleap-support”

Requirements

  • Python 3

Workflow steps

9 steps, taken from the step headings in SKILL.md.

  1. Create Investigation Folder
  2. Fetch Post Content
  3. Download Images
  4. Create USER_POST.md
  5. Write Investigation README
  6. Check Release History First
  7. Analyze and Reproduce
  8. Determine Data Needs
  9. Draft Response

What it can do on your machine

Read from SKILL.md and the folder at commit 4786f14. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • gh
    • git
    • wget
    • uvx
    • python

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • slp.sh

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Sleap Support loads about 2k tokens when it runs. Until then it costs about 85 tokens; SKILL.md has 460 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~85
When it runs · the whole SKILL.md, loaded when a task matches
~2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from talmolab/sleap at commit 4786f14, republished under its BSD-3-Clause-Clear licence (© talmolab). 460 words, ~2,050 tokens.

Download SKILL.mdSave it as .claude/skills/sleap-support/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
sleap-support
description
Handle SLEAP GitHub support workflow for issues and discussions. Use when the user says "support", provides a GitHub issue/discussion number like "#2512", or asks to investigate a user report from talmolab/sleap. Scaffolds investigation folders, downloads posts with images, analyzes problems, and drafts friendly responses.

SLEAP Support Workflow

Handle GitHub issues and discussions from talmolab/sleap with a systematic investigation process.

Quick Start

When given an issue/discussion number:

bash
# Check both issues AND discussions (users often post in wrong category)
gh issue view 2512 --repo talmolab/sleap --json number,title,body,author,createdAt,comments 2>/dev/null || \
gh api repos/talmolab/sleap/discussions/2512 2>/dev/null

Workflow Steps

1. Create Investigation Folder

First, check if an investigation already exists:

bash
ls -d scratch/*-*2512* 2>/dev/null || echo "No existing investigation"

If none exists, create one:

bash
mkdir -p scratch/$(date +%Y-%m-%d)-{issue|discussion}-2512-{short-description}
2. Fetch Post Content

For Issues:

bash
gh issue view NUMBER --repo talmolab/sleap --json number,title,body,author,createdAt,comments,labels > scratch/.../issue.json

For Discussions:

bash
gh api repos/talmolab/sleap/discussions/NUMBER > scratch/.../discussion.json
3. Download Images

Extract image URLs from the post body and download:

bash
# Parse markdown image links: ![alt](url)
grep -oP '!\[.*?\]\(\K[^)]+' scratch/.../issue.json | while read url; do
  wget -P scratch/.../images/ "$url"
done
4. Create USER_POST.md

Convert the JSON to readable markdown with inline images:

markdown
# Issue/Discussion #NUMBER: Title

**Author**: @username
**Created**: YYYY-MM-DD
**Platform**: (extract from post if mentioned)
**SLEAP Version**: (extract from post if mentioned)

## Original Post

[post body with images referenced inline]

## Comments

[any replies]
5. Write Investigation README

Create scratch/.../README.md:

markdown
# Investigation: Issue/Discussion #NUMBER

**Date**: YYYY-MM-DD
**Post**: https://github.com/talmolab/sleap/{issues|discussions}/NUMBER
**Author**: @username
**Type**: Bug Report | Usage Question | Feature Request

## Summary

[1-2 sentence summary of the issue]

## Key Information

- **Platform**: Windows/macOS/Linux
- **SLEAP Version**: X.Y.Z
- **GPU**: (if relevant)
- **Dataset**: (if described)

## Preliminary Analysis

[Initial thoughts on what might be happening]

## Areas to Investigate

- [ ] Check area 1
- [ ] Check area 2

## Files

- `USER_POST.md` - Original post content
- `images/` - Downloaded screenshots
- `RESPONSE_DRAFT.md` - Draft response (when ready)
6. Check Release History First

Before deep investigation, check if the issue was already fixed:

bash
# Get user's SLEAP version from their post (look for sleap doctor output or sleap.__version__)
USER_VERSION="1.4.0"  # example

# Check sleap releases for fixes
gh release list --repo talmolab/sleap --limit 20
gh release view v1.5.0 --repo talmolab/sleap --json body -q '.body' | grep -i "fix"

# For sleap-io issues
gh release list --repo talmolab/sleap-io --limit 10
gh release view v0.6.0 --repo talmolab/sleap-io --json body -q '.body'

# For sleap-nn/training issues
gh release list --repo talmolab/sleap-nn --limit 10
gh release view v0.2.0 --repo talmolab/sleap-nn --json body -q '.body'

If a fix exists in a newer version:

  • Response should guide user to upgrade
  • Include the specific version with the fix
  • Mention what was fixed (link to PR/issue if available)

If investigating an unfixed bug:

  • Checkout the user's version to see their actual code:
bash
# Clone repos if not present
[ -d scratch/repos/sleap-io ] || gh repo clone talmolab/sleap-io scratch/repos/sleap-io
[ -d scratch/repos/sleap-nn ] || gh repo clone talmolab/sleap-nn scratch/repos/sleap-nn

# Checkout user's version
cd scratch/repos/sleap-io && git fetch --tags && git checkout v0.5.0
cd scratch/repos/sleap-nn && git fetch --tags && git checkout v0.1.5

# Now you're looking at the code they're actually running

Use git blame to find potential culprits:

bash
# Find when a suspicious function was last changed
git blame -L 50,100 sleap/io/main.py

# Check if a line was changed recently
git log --oneline -5 -- path/to/file.py

# Find the commit that introduced a specific change
git log -S "function_name" --oneline
7. Analyze and Reproduce

Determine the issue type:

Usage Question: Check if documentation covers this. Common topics:

  • Model configuration (skeleton, training params)
  • Multi-animal vs single-animal tracking
  • Inference and tracking settings
  • Data format questions

Bug Report: Try to reproduce. Check:

  • Version-specific issues
  • Platform-specific behavior
  • GPU/CUDA compatibility
  • Data corruption signs

Feature Request: Note for tracking, no immediate action needed.

8. Determine Data Needs

When to request SLP file:

  • Inference/tracking issues that can't be diagnosed from logs
  • "Labels not showing" or display issues
  • Merging or import problems
  • Corruption or data loss reports

Suggest upload to https://slp.sh - our SLP file sharing service.

If SLP provided: Download and analyze:

bash
sio show path/to/file.slp --summary
sio show path/to/file.slp --videos
sio show path/to/file.slp --skeleton
9. Draft Response

Create RESPONSE_DRAFT.md following this structure:

markdown
Hi @{username},

Thanks for the post!

[Restate understanding: "If I understand correctly, you're seeing X when you try to Y..."]

[Provide solution OR request more info]

[If requesting info, give EXPLICIT instructions:]
- Use `sleap doctor` CLI for diagnostics
- Provide copy-paste terminal commands
- Assume non-technical user

Let us know if that works for you!

Cheers,

:heart: Talmo & Claude :robot:

<details>
<summary><b>Extended technical analysis</b></summary>

[Detailed investigation notes, code traces, version checks]

</details>

Response Tone Guidelines

  • Opening: Bright and positive ("Thanks for the post!", "Great question!")
  • Body: Clear, concise, non-technical language
  • Instructions: Step-by-step, assume terminal newbie
  • Closing: Encouraging ("Let us know if that works for you!")
  • Signature: :heart: Talmo & Claude :robot:
Show full SKILL.md (166 more words)Show less

When Requesting User Actions

Terminal commands must be copy-paste ready:

bash
# Good - one-liner, no environment activation needed
sleap doctor

# Good - uses uvx for isolated execution
uvx sio show your_file.slp --summary

# Bad - assumes environment knowledge
source activate sleap && python -c "import sleap; print(sleap.__version__)"

For data/I/O issues - Check talmolab/sleap-io:

  • Local: ../sleap-io (preferred)
  • Clone if needed: gh repo clone talmolab/sleap-io scratch/repos/sleap-io
  • Key docs: sleap-io/docs/examples.md, sleap-io/docs/formats/SLP.md
  • CLI: sio show --help for inspection commands

For training/inference issues - Check talmolab/sleap-nn:

  • Local: ../sleap-nn (preferred)
  • Clone if needed: gh repo clone talmolab/sleap-nn scratch/repos/sleap-nn
  • Topics: Model configs, training params, evaluation metrics, tracking

Common Patterns

Instance Duplication
  • Check track assignment logic
  • Look for ID switching during tracking
  • May need sleap-track with different settings
Training Issues
  • GPU memory: suggest reducing batch size
  • Loss not decreasing: check learning rate, augmentation
  • NaN losses: data normalization issues
Import/Export Issues
  • Format compatibility (H5 vs SLP versions)
  • Missing video paths
  • Skeleton definition mismatches
GUI Issues
  • Qt/PySide6 version conflicts
  • Display scaling on high-DPI
  • Video codec issues

Confirmation Before Posting

ALWAYS confirm with the developer before posting:

  1. Show the full draft response
  2. Ask: "Does this look good to post?"
  3. Wait for explicit approval

Never post automatically - support responses represent the project.

© talmolab, BSD-3-Clause-Clear. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files in .claude/skills/sleap-support of talmolab/sleap.

  • SKILL.md
  • gh-commands.md
  • response-templates.md
  • troubleshooting-patterns.md

Open the folder on GitHubat commit 4786f14

Compare with similar skills

Sleap Support next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Sleap Support compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Sleap Support this skilltalmolab/sleap618—~2kAutomated safety check: PassBSD-3-Clause-Clear
degit Project ScaffoldingRich-Harris/degit7.9k—~534Automated safety check: PassMIT
Rust Hygiene Audittsz-org/tsz577—~1.5kAutomated safety check: PassApache-2.0
Create Vechain Dappvechain/x-app-template450—~1.8kAutomated safety check: PassMIT
Upgrade Starter Kitworkadventure/map-starter-kit156—~1.3kAutomated safety check: NotesCustom licence
Create PRgenlayerlabs/genlayer-studio180—~635Automated safety check: PassMIT

Similar skills

  • degit Project Scaffolding

    Rich-Harris/degit

    Downloads a repository snapshot or template with degit into an empty folder, from GitHub, GitLab, Bitbucket, Sourcehut or a Gist, optionally at a branch, tag or commit.

    7.9k GitHub stars~534 tokensUpdated 26 days ago
    DevelopmentAuto-check passed
  • Run a deep DRY + code-hygiene audit of the Rust workspace and turn the findings into verified, deduplicated, hierarchical GitHub tech-debt issues.

    577 GitHub stars~1.5k tokensUpdated 1 mo ago
    DevelopmentAuto-check passed
  • Create Vechain Dapp

    vechain/x-app-template

    Scaffold a VeChain dApp with Next.js, VeChain Kit, Chakra UI v3, and GitHub Pages deployment.

    450 GitHub stars~1.8k tokensUpdated 2 mo ago
    DevelopmentAuto-check passed
  • Upgrade Starter Kit

    workadventure/map-starter-kit

    Upgrade a WorkAdventure map repository to the latest version of the map-starter-kit (github.com/workadventure/map-starter-kit) - refreshes package.json dependencies, vite/tsconfig/build config, CI…

    156 GitHub stars~1.3k tokensUpdated 7 days ago
    DevelopmentAuto-check: notes
  • Create PR

    genlayerlabs/genlayer-studio

    Create GitHub pull requests by analyzing branch diff and following the project template

    180 GitHub stars~635 tokensUpdated yesterday
    DevelopmentAuto-check passed
  • Project Initialization

    pavel-molyanov/molyanov-ai-dev

    Initializes a project from the standard dual-runtime template, preserves existing files, configures Git hooks, and creates or connects a private GitHub repository with main and dev branches.

    297 GitHub stars~1.1k tokensUpdated 1 mo ago
    DevelopmentAuto-check: notes

More from talmolab/sleap

  • Qt Testing

    talmolab/sleap

    Capture and visually inspect Qt GUI widgets using screenshots.

    618 GitHub stars~557 tokensUpdated 9 days ago
    Auto-check passed
  • Investigation

    talmolab/sleap

    Scaffolds a structured investigation in scratch/ for empirical research and documentation.

    618 GitHub stars~899 tokensUpdated 9 days ago
    Auto-check passed
  • PR

    talmolab/sleap

    Create a well-structured GitHub PR with proper branching, testing, formatting, and documentation.

    618 GitHub stars~1.7k tokensUpdated 9 days ago
    Auto-check passed

Works with

Categories

Questions about Sleap Support

What does Sleap Support do?

Handle SLEAP GitHub support workflow for issues and discussions. Sleap Support is an agent skill from talmolab/sleap. Handle SLEAP GitHub support workflow for issues and discussions.

When should I use Sleap Support?

Sleap Support fits situations like: the user says support; provides a GitHub issue/discussion number like 2512; asks to investigate a user report from talmolab/sleap.

How do I install Sleap Support in Claude Code?

Run `npx skills add talmolab/sleap --skill sleap-support -a claude-code`. Or copy the skill folder (.claude/skills/sleap-support in talmolab/sleap) into .claude/skills/sleap-support in your project. Claude Code loads it when a task matches its description.

How do I install Sleap Support in Codex?

Run `npx skills add talmolab/sleap --skill sleap-support -a codex`. Or copy the skill folder (.claude/skills/sleap-support in talmolab/sleap) into .agents/skills/sleap-support in your project. Codex loads it when a task matches its description.

Can I use Sleap Support in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add talmolab/sleap --skill sleap-support -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/sleap-support, .gemini/skills/sleap-support, .github/skills/sleap-support and .opencode/skills/sleap-support in your project.

What does Sleap Support need to run?

Going by SKILL.md and its folder, Sleap Support needs the command-line tools its instructions call (gh, git, wget, uvx and python). Our summary lists: Python 3.

Does Sleap Support access the network?

SKILL.md names 1 domain. In commands or code: slp.sh; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.

Is Sleap Support safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Sleap Support use?

Sleap Support is published under the BSD-3-Clause-Clear licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Sleap Support use?

About 2k tokens (SKILL.md is roughly 8.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Sleap Support?

Skills that share tags, products or a category with Sleap Support: degit Project Scaffolding (Rich-Harris/degit, 7.9k stars), Rust Hygiene Audit (tsz-org/tsz, 577 stars), Create Vechain Dapp (vechain/x-app-template, 450 stars) and Upgrade Starter Kit (workadventure/map-starter-kit, 156 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Sleap Support?

talmolab (a GitHub organization) maintains it in talmolab/sleap, which has 618 GitHub stars. The repository holds 4 skills in this directory. The repository was last updated on September 30, 2026.

Source: talmolab/sleap on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.