A collection of open-source skills for AI coding agents (Claude Code, OpenClaw, etc.) powered by the Pika Developer API.
Skills are self-contained modules that extend the capabilities of AI coding agents. Each skill is a directory containing:
SKILL.md— The skill definition file. It tells the AI agent when to activate the skill, how to use it, and includes step-by-step instructions. The agent reads this file and follows the workflow described in it.scripts/— Executable scripts (Python, Bash, etc.) that the agent invokes as part of the skill workflow.requirements.txt— Python dependencies needed by the skill scripts.
When you install a skill into your agent workspace, the agent automatically detects the SKILL.md and knows how to use the skill — no manual configuration needed.
| Skill | Pricing | Description |
|---|---|---|
| pikastream-video-meeting | $0.275 / min | Join a Google Meet as a real-time AI avatar. |
Go to https://www.pika.me/dev/ and create a Developer Key (starts with dk_).
export PIKA_DEV_KEY="dk_your-key-here"Or add it to your shell profile (~/.bashrc, ~/.zshrc, etc.) for persistence.
Point your agent to the skill folder and ask it to install. For example:
install /path/to/your/Pika-Skills/pikastream-video-meeting/
Once installed, simply interact with your AI agent naturally. For example, drop a Google Meet link and the agent will automatically activate the pikastream-video-meeting skill.
The skill will check your balance before joining. If you need credits, it will generate a payment link for you automatically.
An AI-powered video meeting bot that joins Google Meet calls as a real-time avatar.
- Real-time avatar — Joins meetings with a generated or custom avatar image via PikaStreaming.
- Voice cloning — Clone your voice from a short audio recording.
- Avatar generation — Generate an AI avatar via OpenAI image models, or provide your own image.
- Automatic billing — Checks balance before joining, creates a payment link if needed, and waits for payment to complete.
- Context-aware conversation — The bot synthesizes workspace context (identity, recent activity, known people) into a system prompt for natural, informed conversation during meetings.
- Post-meeting notes — Automatically retrieves and shares meeting notes after the bot leaves.
The skill script supports four subcommands:
# Join a Google Meet
python scripts/pikastreaming_videomeeting.py join \
--meet-url <google-meet-url> --bot-name <name> \
--image <avatar-image> [--voice-id <id>] [--system-prompt-file <path>]
# Leave a meeting
python scripts/pikastreaming_videomeeting.py leave --session-id <id>
# Generate an avatar image
python scripts/pikastreaming_videomeeting.py generate-avatar --output <path> [--prompt <text>]
# Clone a voice from audio
python scripts/pikastreaming_videomeeting.py clone-voice \
--audio <file> --name <name> [--noise-reduction]| Variable | Required | Description |
|---|---|---|
PIKA_DEV_KEY |
Yes | Pika Developer Key (dk_...). Get one at pika.me/dev |
- Python 3.10+
PIKA_DEV_KEYenvironment variableffmpeg(optional, for audio format conversion during voice cloning)
To add a new skill:
- Create a new directory with your skill name.
- Add a
SKILL.mdfollowing the frontmatter format (see existing skills for reference). - Add scripts and a
requirements.txtas needed. - Update this README with your skill's description.
Apache 2.0 — see LICENSE.