skywork-ppt
Generate PPTs from topics or templates, edit existing presentations via natural language, or perform local file operations (delete/reorder/merge slides). Trigger on requests to create, modify, or manipulate .pptx files. Built on Nano Banana 2. Requires Python 3.8+.
What this skill does
# PPT Write Skill
Four capabilities: **generate**, **template imitation**, **edit existing PPT**, and **local file operations**.
---
## Authentication (Required First)
Before using this skill, authentication must be completed. Run the auth script first:
```bash
# Authenticate: checks env token / cached token / browser login
python3 <skill-dir>/scripts/skywork_auth.py || exit 1
```
**Token priority**:
1. Environment variable `SKYBOT_TOKEN` → if set, use directly
2. Cached token file `~/.skywork_token` → validate via API, if valid, use it
3. No valid token → opens browser for login, polls until complete, saves token
**IMPORTANT - Login URL handling**: If script output contains a line starting with `[LOGIN_URL]`, you **MUST** immediately send that URL to the user in a clickable message (e.g. "Please open this link to log in: <url>"). The user may be in an environment where the browser cannot open automatically, so always surface the login URL.
---
## Routing — Identify the user's intent first
| User intent | Which path |
|-------------|------------|
| Generate a new PPT from a topic, set of requirements or reference files | **Layer 1** — Generate |
| Use an existing .pptx as a layout/style template to create a new presentation | **Layer 2** — Imitate |
| Edit an existing PPT: modify slides, add slides, change style, split/merge | **Layer 4** — Edit |
| Delete / reorder / extract / merge slides in a local file (no backend) | **Layer 3** — Local ops |
---
## Environment check (always run this first)
**This skill requires Python 3 (>=3.8). Run the following before any script to locate a valid Python binary and install dependencies.**
```bash
PYTHON_CMD=""
for cmd in python3 python python3.13 python3.12 python3.11 python3.10 python3.9 python3.8; do
if command -v "$cmd" &>/dev/null && "$cmd" -c "import sys; exit(0 if sys.version_info >= (3,8) else 1)" 2>/dev/null; then
PYTHON_CMD="$cmd"
break
fi
done
if [ -z "$PYTHON_CMD" ]; then
echo "ERROR: Python 3.8+ not found."
echo "Install on macOS: brew install python3 or visit https://www.python.org/downloads/"
exit 1
fi
echo "Found Python: $PYTHON_CMD ($($PYTHON_CMD --version))"
$PYTHON_CMD -m pip install -q --break-system-packages python-pptx
echo "Dependencies ready."
```
> After this check, replace `python` with the discovered `$PYTHON_CMD` (e.g. `python3`) in all subsequent commands.
---
## Layer 1 — Generate PPT
### Steps
0. **REQUIRED FIRST STEP** — Read [workflow_generate.md](workflow_generate.md) NOW, before taking any other action. After reading, output exactly: `✅ workflow_generate.md loaded.` — then proceed.
1. **Environment check** — run the check above to get `$PYTHON_CMD`.
2. **Upload reference files** (if the user provides local files as content source) — parse the file using tool in script/parse_file.py and pass the result to `--files`. See the `--files` note below.
3. **Web search** (required if no relevant content is already in the conversation) — call web_search tool in script to search the topic and distill results into a `reference-file` file of ≤ 2000 words.
4. **Run the script**:
> **Important**: set exec tool `yieldMs` to `600000` (10 minutes).
5. **Deliver** — provide the absolute `.pptx` path and the download URL.
---
## Layer 2 — Imitate PPT (template-based generation)
### Steps
0. **REQUIRED FIRST STEP** - Read [workflow_imitate.md](workflow_imitate.md) immidiately before any action you do!!!
1. **Environment check** — run the check above to get `$PYTHON_CMD`.
2. **Locate the template** — extract the absolute path of the local `.pptx` from the user's message; ask the user if it's unclear.
3. **Upload the template** — upload it and extract `TEMPLATE_URL` from the output.
4. **Upload reference files** (if the user provides additional local files as content source) — parse the file using tool in script/parse_file.py and pass the result to `--files`. See the `--files`
5. **Web search** (required if no relevant content is already in the conversation) — call web_search tool in script to search the new topic and distill results into a `reference-file` file of ≤ 2000 words.
6. **Run the script**:
> **Important**: set exec tool `yieldMs` to `600000` (10 minutes).
7. **Deliver** — provide the absolute `.pptx` path, the download URL, and the template filename used.
---
## Layer 4 — Edit PPT (AI-powered modification)
Use this layer when the user wants to modify an existing PPT using natural language. Requires an OSS/CDN URL of the PPTX (from a previous generation or upload).
### Steps
0. **Detailed workflow** - Read [workflow_edit.md](workflow_edit.md) immediately before any action you do!!!
1. **Environment check** — run the check above to get `$PYTHON_CMD`.
2. **Get PPTX URL** — from the user's message or upload a local file first.
3. **Run the script** with `--pptx-url`:
```bash
$PYTHON_CMD scripts/run_ppt_write.py "edit instruction" \
--language Chinese \
--pptx-url "https://cdn.example.com/file.pptx" \
-o /absolute/path/output.pptx
```
> **Important**: set exec tool `yieldMs` to `600000` (10 minutes).
4. **Deliver** — provide download link, local path, and summary of changes.
---
## Layer 3 — Local file operations
```bash
# Inspect slide count and titles
$PYTHON_CMD scripts/local_pptx_ops.py info --file my.pptx
# Delete slides (1-based index; supports ranges like 3,5,7-9; omit -o to overwrite in place)
$PYTHON_CMD scripts/local_pptx_ops.py delete --file my.pptx --slides 3,5,7-9 -o trimmed.pptx
# Reorder slides (must list every slide, no omissions)
$PYTHON_CMD scripts/local_pptx_ops.py reorder --file my.pptx --order 2,1,4,3,5
# Extract a subset of slides into a new file
$PYTHON_CMD scripts/local_pptx_ops.py extract --file my.pptx --slides 1-3 -o subset.pptx
# Merge multiple files
$PYTHON_CMD scripts/local_pptx_ops.py merge --files a.pptx b.pptx -o merged.pptx
```
Read [workflow_local.md](workflow_local.md) immidiately before any action you do!!!
---
## Error Handling
- **Insufficient benefit**: When calling scripts (generate, imitate, or edit), the script or log may show a message like `Insufficient benefit. Please upgrade your account at {url}`, meaning the user's benefit level does not meet the requirement for this skill.
### How to reply when benefit is insufficient
When you detect the above, **reply in the user's current language** — do not echo the English message. Use this pattern:
- Convey: "Sorry, PPT generation failed. This skill requires upgrading your Skywork membership to use." then a single call-to-action link.
- **Format**: One short sentence in the user's language + a link like `[Upgrade now →](url)` or the equivalent in their language .
- **URL**: Extract the upgrade URL from the log/script output (e.g. the `at https://...` part).
---
## Dependencies
- **Python 3.8+** (required) — `python3` / `python` must be on PATH
- Layer 3 local ops: `pip install python-pptx --break-system-packages`
(The environment check step installs all required dependencies automatically.)
---
## Which layer to trigger?
| Scenario | Use |
|----------|-----|
| Generate a PPT from a topic or existing reference files | Layer 1 |
| Imitate the layout/style of an existing .pptx | Layer 2 |
| Edit/modify an existing PPT via natural language | Layer 4 |
| Delete / reorder / extract / merge local .pptx files (no backend) | Layer 3 |
Related in Image & Video
watch
IncludedWatch a video (URL or local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or Whisper API fallback), and hands the result to Claude so it can answer questions about what's in the video.
physical-ai-defect-image-generation
IncludedUse when the user wants to orchestrate defect image generation, run associated setup, or handle outputs on OSMO. The Day 0 path handles cold-start with USD-to-ROI, image-edit augmentation, and AnomalyGen to create initial PCBA datasets. The Day 1 path performs inference and labeling on real images. This skill helps with first-time asset setup, creation of finetuning checkpoints, and configuring deployment. Trigger keywords: defect image generation, dig workflow, dig pipeline, defect image detection workflow, aoi pipeline, aoi anomalygen, usd2roi anomalygen, day 0 pcba, day 1 pcba, day 1 real-photo alignment, day 1 manual roi, metal surface anomaly, glass defect, anomalygen finetune, setup_pcb, setup_metal, setup_glass, setup_pretrained, dig setup, dig datasets, dig pretrained checkpoint, dig image-edit endpoint.
accelint-react-best-practices
IncludedReact performance optimization and best practices. ALWAYS use this skill when working with any React code - writing components, hooks, JSX; refactoring; optimizing re-renders, memoization, state management; reviewing for performance; fixing hydration mismatches; debugging infinite re-renders, stale closures, input focus loss, animations restarting; preventing remounting; implementing transitions, lazy initialization, effect dependencies. Even simple React tasks benefit from these patterns. Covers React 19+ (useEffectEvent, Activity, ref props). Triggers - useEffect, useState, useMemo, useCallback, memo, inline components, nested components, components inside components, re-render, performance, hydration, SSR, Next.js, useDeferredValue, combined hooks.
elevenlabs-agents
IncludedBuild conversational AI voice agents with ElevenLabs Platform using React, JavaScript, React Native, or Swift SDKs. Configure agents, tools (client/server/MCP), RAG knowledge bases, multi-voice, and Scribe real-time STT. Use when: building voice chat interfaces, implementing AI phone agents with Twilio, configuring agent workflows or tools, adding RAG knowledge bases, testing with CLI "agents as code", or troubleshooting deprecated @11labs packages, Android audio cutoff, CSP violations, dynamic variables, or WebRTC config. Keywords: ElevenLabs Agents, ElevenLabs voice agents, AI voice agents, conversational AI, @elevenlabs/react, @elevenlabs/client, @elevenlabs/react-native, @elevenlabs/elevenlabs-js, @elevenlabs/agents-cli, elevenlabs SDK, voice AI, TTS, text-to-speech, ASR, speech recognition, turn-taking model, WebRTC voice, WebSocket voice, ElevenLabs conversation, agent system prompt, agent tools, agent knowledge base, RAG voice agents, multi-voice agents, pronunciation dictionary, voice speed control, elevenlabs scribe, @11labs deprecated, Android audio cutoff, CSP violation elevenlabs, dynamic variables elevenlabs, case-sensitive tool names, webhook authentication
humanizer
IncludedHumanize AI-generated text by detecting and removing patterns typical of LLM output. Rewrites text to sound natural, specific, and human. Uses 28 pattern detectors, 560+ AI vocabulary terms across 3 tiers, and statistical analysis (burstiness, type-token ratio, readability) for comprehensive detection. Use when asked to humanize text, de-AI writing, make content sound more natural/human, review writing for AI patterns, score text for AI detection, or improve AI-generated drafts. Covers content, language, style, communication, and filler categories.
generating-mermaid-diagrams
IncludedSalesforce architecture diagrams using Mermaid with ASCII fallback. Use this skill when generating text-based diagrams for Salesforce architecture, OAuth flows, ERDs, integration sequences, or Agentforce structure. TRIGGER when: user says "diagram", "visualize", "ERD", or asks for sequence diagrams, flowcharts, class diagrams, or architecture visualizations in Mermaid. DO NOT TRIGGER when: user wants PNG/SVG image output (use generating-visual-diagrams), or asks about non-Salesforce systems.