pipecat-friday-agent — Iron Man-inspired tactical voice assistant (F.R.I.D.A.Y.) with Pipecat, Gemini, and OpenAI. (#261)
* feat: add pipecat-friday-agent skill and sync generated files * chore: sync generated files and update changelog for pipecat-friday-agent * fix: normalize path separators in data registry for cross-platform CI * chore: sync generated registry files and resolve CI drift * feat: Add initial skill catalog, aliases, and bundles data files and update README. * fix: normalize path separators in skill utils and resolve CI drift * feat: Introduce skill catalog, aliases, and bundles with a new utility and updated documentation. * feat: Add new skill catalog data files for aliases, bundles, and the main catalog, and update total skill count. * feat: Update skill and star counts, and registry sync metadata in README.md. * chore: sync generated registry files * chore: fix drift --------- Co-authored-by: sck_0 <samujackson1337@gmail.com>
This commit is contained in:
Takura
GitHub
@@ -2,7 +2,7 @@
|
||||
|
||||
Generated at: 2026-02-08T00:00:00.000Z
|
||||
|
||||
Total skills: 1236
|
||||
Total skills: 1237
|
||||
|
||||
## architecture (80)
|
||||
|
||||
@@ -408,7 +408,7 @@ Scope::with_data, save state, load state, serde,
|
||||
| `yann-lecun` | Agente que simula Yann LeCun — inventor das Convolutional Neural Networks, Chief AI Scientist da Meta, Prêmio Turing 2018. Use quando quiser: perspectivas so... | persona, cnn, meta, ai-safety-critic, open-source | persona, cnn, meta, ai-safety-critic, open-source, yann, lecun, agente, que, simula, inventor, das |
|
||||
| `youtube-automation` | Automate YouTube tasks via Rube MCP (Composio): upload videos, manage playlists, search content, get analytics, and handle comments. Always search tools firs... | youtube | youtube, automation, automate, tasks, via, rube, mcp, composio, upload, videos, playlists, search |
|
||||
|
||||
## development (178)
|
||||
## development (179)
|
||||
|
||||
| Skill | Description | Tags | Triggers |
|
||||
| --- | --- | --- | --- |
|
||||
@@ -547,6 +547,7 @@ no matching field, parse error, widget... | makepad, reference | makepad, refere
|
||||
| `odoo-shopify-integration` | Connect Odoo with Shopify: sync products, inventory, orders, and customers using the Shopify API and Odoo's external API or connector modules. | odoo, shopify, integration | odoo, shopify, integration, connect, sync, products, inventory, orders, customers, api, external, connector |
|
||||
| `odoo-woocommerce-bridge` | Sync Odoo with WooCommerce: products, inventory, orders, and customers via WooCommerce REST API and Odoo external API. | odoo, woocommerce, bridge | odoo, woocommerce, bridge, sync, products, inventory, orders, customers, via, rest, api, external |
|
||||
| `pakistan-payments-stack` | Design and implement production-grade Pakistani payment integrations (JazzCash, Easypaisa, bank/PSP rails, optional Raast) for SaaS with PKR billing, webhook... | saas, payments, pakistan, nextjs, b2b, pkr, reconciliation | saas, payments, pakistan, nextjs, b2b, pkr, reconciliation, stack, grade, pakistani, payment, integrations |
|
||||
| `pipecat-friday-agent` | Build a low-latency, Iron Man-inspired tactical voice assistant (F.R.I.D.A.Y.) using Pipecat, Gemini, and OpenAI. | pipecat, voice, gemini, openai, python | pipecat, voice, gemini, openai, python, friday, agent, low, latency, iron, man, inspired |
|
||||
| `playwright-java` | Scaffold, write, debug, and enhance enterprise-grade Playwright E2E tests in Java using Page Object Model, JUnit 5, Allure reporting, and parallel execution. | playwright, java, e2e-testing, junit5, page-object-model, allure, selenium-alternative | playwright, java, e2e-testing, junit5, page-object-model, allure, selenium-alternative, scaffold, write, debug, enhance, enterprise |
|
||||
| `product-manager-toolkit` | Comprehensive toolkit for product managers including RICE prioritization, customer interview analysis, PRD templates, discovery frameworks, and go-to-market ... | product, manager | product, manager, toolkit, managers, including, rice, prioritization, customer, interview, analysis, prd, discovery |
|
||||
| `python-development-python-scaffold` | You are a Python project architecture expert specializing in scaffolding production-ready Python applications. Generate complete project structures with mode... | python | python, development, scaffold, architecture, specializing, scaffolding, applications, generate, complete, structures, tooling, uv |
|
||||
|
||||
@@ -9,6 +9,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
|
||||
|
||||
## [Unreleased]
|
||||
|
||||
- **pipecat-friday-agent** — Iron Man-inspired tactical voice assistant (F.R.I.D.A.Y.) with Pipecat, Gemini, and OpenAI.
|
||||
|
||||
---
|
||||
|
||||
## [7.4.0] - 2026-03-10 - "Planning & Dashboards"
|
||||
|
||||
12
README.md
12
README.md
@@ -1,7 +1,7 @@
|
||||
<!-- registry-sync: version=7.4.0; skills=1236; stars=22821; updated_at=2026-03-10T16:36:51+00:00 -->
|
||||
# 🌌 Antigravity Awesome Skills: 1,236+ Agentic Skills for Claude Code, Gemini CLI, Cursor, Copilot & More
|
||||
<!-- registry-sync: version=7.4.0; skills=1237; stars=22850; updated_at=2026-03-10T18:13:31+00:00 -->
|
||||
# 🌌 Antigravity Awesome Skills: 1,237+ Agentic Skills for Claude Code, Gemini CLI, Cursor, Copilot & More
|
||||
|
||||
> **The Ultimate Collection of 1,236+ Universal Agentic Skills for AI Coding Assistants — Claude Code, Gemini CLI, Codex CLI, Antigravity IDE, GitHub Copilot, Cursor, OpenCode, AdaL**
|
||||
> **The Ultimate Collection of 1,237+ Universal Agentic Skills for AI Coding Assistants — Claude Code, Gemini CLI, Codex CLI, Antigravity IDE, GitHub Copilot, Cursor, OpenCode, AdaL**
|
||||
|
||||
[](https://github.com/sickn33/antigravity-awesome-skills/stargazers)
|
||||
[](LICENSE)
|
||||
@@ -18,7 +18,7 @@
|
||||
[](apps/web-app)
|
||||
[](https://buymeacoffee.com/sickn33)
|
||||
|
||||
**Antigravity Awesome Skills** is a curated, battle-tested library of **1,236+ high-performance agentic skills** designed to work seamlessly across the major AI coding assistants.
|
||||
**Antigravity Awesome Skills** is a curated, battle-tested library of **1,237+ high-performance agentic skills** designed to work seamlessly across the major AI coding assistants.
|
||||
|
||||
**Welcome to the V7.4.0 Release!** This repository gives your agent reusable playbooks for planning, coding, debugging, testing, security review, infrastructure work, product thinking, and much more.
|
||||
|
||||
@@ -32,7 +32,7 @@
|
||||
- [🎁 Curated Collections (Bundles)](#curated-collections)
|
||||
- [🧭 Antigravity Workflows](#antigravity-workflows)
|
||||
- [📦 Features & Categories](#features--categories)
|
||||
- [📚 Browse 1,236+ Skills](#browse-1236-skills)
|
||||
- [📚 Browse 1,237+ Skills](#browse-1237-skills)
|
||||
- [🤝 How to Contribute](#how-to-contribute)
|
||||
- [💬 Community](#community)
|
||||
- [☕ Support the Project](#support-the-project)
|
||||
@@ -281,7 +281,7 @@ The repository is organized into specialized domains to transform your AI into a
|
||||
|
||||
Counts change as new skills are added. For the current full registry, see [CATALOG.md](CATALOG.md).
|
||||
|
||||
## Browse 1,236+ Skills
|
||||
## Browse 1,237+ Skills
|
||||
|
||||
- Open the interactive browser in [`apps/web-app`](apps/web-app).
|
||||
- Read the full catalog in [`CATALOG.md`](CATALOG.md).
|
||||
|
||||
@@ -237,6 +237,7 @@
|
||||
"openapi-spec-generation",
|
||||
"pakistan-payments-stack",
|
||||
"php-pro",
|
||||
"pipecat-friday-agent",
|
||||
"plaid-fintech",
|
||||
"playwright-java",
|
||||
"polars",
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
{
|
||||
"generatedAt": "2026-02-08T00:00:00.000Z",
|
||||
"total": 1236,
|
||||
"total": 1237,
|
||||
"skills": [
|
||||
{
|
||||
"id": "00-andruia-consultant",
|
||||
@@ -21598,6 +21598,34 @@
|
||||
],
|
||||
"path": "skills/php-pro/SKILL.md"
|
||||
},
|
||||
{
|
||||
"id": "pipecat-friday-agent",
|
||||
"name": "pipecat-friday-agent",
|
||||
"description": "Build a low-latency, Iron Man-inspired tactical voice assistant (F.R.I.D.A.Y.) using Pipecat, Gemini, and OpenAI.",
|
||||
"category": "development",
|
||||
"tags": [
|
||||
"pipecat",
|
||||
"voice",
|
||||
"gemini",
|
||||
"openai",
|
||||
"python"
|
||||
],
|
||||
"triggers": [
|
||||
"pipecat",
|
||||
"voice",
|
||||
"gemini",
|
||||
"openai",
|
||||
"python",
|
||||
"friday",
|
||||
"agent",
|
||||
"low",
|
||||
"latency",
|
||||
"iron",
|
||||
"man",
|
||||
"inspired"
|
||||
],
|
||||
"path": "skills/pipecat-friday-agent/SKILL.md"
|
||||
},
|
||||
{
|
||||
"id": "pipedrive-automation",
|
||||
"name": "pipedrive-automation",
|
||||
|
||||
79
skills/pipecat-friday-agent/SKILL.md
Normal file
79
skills/pipecat-friday-agent/SKILL.md
Normal file
@@ -0,0 +1,79 @@
|
||||
---
|
||||
name: pipecat-friday-agent
|
||||
description: "Build a low-latency, Iron Man-inspired tactical voice assistant (F.R.I.D.A.Y.) using Pipecat, Gemini, and OpenAI."
|
||||
category: voice-agents
|
||||
risk: safe
|
||||
source: community
|
||||
date_added: "2026-03-10"
|
||||
tags: [pipecat, voice, gemini, openai, python]
|
||||
tools: [pipecat]
|
||||
---
|
||||
|
||||
# Pipecat Friday Agent
|
||||
|
||||
## Overview
|
||||
|
||||
This skill provides a blueprint for building **F.R.I.D.A.Y.** (Replacement Integrated Digital Assistant Youth), a local voice assistant inspired by the tactical AI from the Iron Man films. It uses the **Pipecat** framework to orchestrate a low-latency pipeline:
|
||||
- **STT**: OpenAI Whisper (`whisper-1`) or `gpt-4o-transcribe`
|
||||
- **LLM**: Google Gemini 2.5 Flash (via a compatibility shim)
|
||||
- **TTS**: OpenAI TTS (`nova` voice)
|
||||
- **Transport**: Local Audio (Hardware Mic/Speakers)
|
||||
|
||||
## When to Use This Skill
|
||||
|
||||
- Use when you want to build a real-time, conversational voice agent.
|
||||
- Use when working with the Pipecat framework for pipeline-based AI.
|
||||
- Use when you need to integrate multiple providers (Google and OpenAI) into a single voice loop.
|
||||
- Use when building Iron Man-themed or tactical-themed voice applications.
|
||||
|
||||
## How It Works
|
||||
|
||||
### Step 1: Install Dependencies
|
||||
|
||||
You will need the Pipecat framework and its service providers installed:
|
||||
```bash
|
||||
pip install pipecat-ai[openai,google,silero] python-dotenv
|
||||
```
|
||||
|
||||
### Step 2: Configure Environment
|
||||
|
||||
Create a `.env` file with your API keys:
|
||||
```env
|
||||
OPENAI_API_KEY=your_openai_key
|
||||
GOOGLE_API_KEY=your_google_key
|
||||
```
|
||||
|
||||
### Step 3: Run the Agent
|
||||
|
||||
Execute the provided Python script to start the interface:
|
||||
```bash
|
||||
python scripts/friday_agent.py
|
||||
```
|
||||
|
||||
## Core Concepts
|
||||
|
||||
### Pipeline Architecture
|
||||
The agent follows a linear pipeline: `Mic -> VAD -> STT -> LLM -> TTS -> Speaker`. This allows for granular control over each stage, unlike end-to-end speech-to-speech models.
|
||||
|
||||
### Google Compatibility Shim
|
||||
Since Google's Gemini API has a different message format than OpenAI's standard (which Pipecat aggregators expect), the script includes a `GoogleSafeContext` and `GoogleSafeMessage` class to bridge the gap.
|
||||
|
||||
## Best Practices
|
||||
|
||||
- ✅ **Use Silero VAD**: It is robust for local hardware and prevents background noise from triggering the LLM.
|
||||
- ✅ **Concise Prompts**: Tactical agents should give short, data-dense responses to minimize latency.
|
||||
- ✅ **Sample Rate Match**: OpenAI TTS outputs at 24kHz; ensure your `audio_out_sample_rate` matches to avoid high-pitched or slowed audio.
|
||||
- ❌ **No Polite Fillers**: Avoid "Hello, how can I help you today?" Instead, use "Systems nominal. Ready for commands."
|
||||
|
||||
## Troubleshooting
|
||||
|
||||
- **Problem:** Audio is choppy or delayed.
|
||||
- **Solution:** Check your `OUTPUT_DEVICE` index. Run a script like `test_audio_output.py` to find the correct hardware index for your OS.
|
||||
- **Problem:** "Validation error" for message format.
|
||||
- **Solution:** Ensure the `GoogleSafeContext` shim is correctly translating OpenAI-style dicts to Gemini-style schema.
|
||||
|
||||
## Related Skills
|
||||
|
||||
- `@voice-agents` - General principles of voice AI.
|
||||
- `@agent-tool-builder` - Add tools (Search, Lights, etc.) to your Friday agent.
|
||||
- `@llm-architect` - Optimizing the LLM layer.
|
||||
146
skills/pipecat-friday-agent/scripts/friday_agent.py
Normal file
146
skills/pipecat-friday-agent/scripts/friday_agent.py
Normal file
@@ -0,0 +1,146 @@
|
||||
import asyncio
|
||||
import os
|
||||
import sys
|
||||
from dotenv import load_dotenv
|
||||
|
||||
from pipecat.pipeline.pipeline import Pipeline
|
||||
from pipecat.pipeline.runner import PipelineRunner
|
||||
from pipecat.pipeline.task import PipelineTask, PipelineParams
|
||||
from pipecat.services.openai.stt import OpenAISTTService
|
||||
from pipecat.services.openai.tts import OpenAITTSService
|
||||
from pipecat.services.google.llm import GoogleLLMService
|
||||
from pipecat.processors.aggregators.llm_response import LLMUserContextAggregator, LLMAssistantContextAggregator
|
||||
from pipecat.audio.vad.silero import SileroVADAnalyzer
|
||||
from pipecat.transports.local.audio import LocalAudioTransport, LocalAudioTransportParams
|
||||
|
||||
# ── Config ────────────────────────────────────────────────────────────────────
|
||||
# run test_audio_output.py to find your device index
|
||||
# [4] Speaker (Realtek) Windows default speakers
|
||||
# [6] Headphones (soundcore Space One) Bluetooth headphones
|
||||
OUTPUT_DEVICE = 6
|
||||
|
||||
# "whisper-1" (classic) or "gpt-4o-transcribe" (GPT-4o powered, higher accuracy)
|
||||
WHISPER_MODEL = "whisper-1"
|
||||
|
||||
# OpenAI TTS voice — alloy, ash, coral, echo, fable, nova, onyx, sage, shimmer
|
||||
# "nova" is calm and professional; "shimmer" is warm; "onyx" is deep
|
||||
TTS_VOICE = "nova"
|
||||
|
||||
# ── Google compatibility shim ─────────────────────────────────────────────────
|
||||
# Pipecat's context aggregators use OpenAI-style {role, content} messages,
|
||||
# but GoogleLLMService expects {role, parts: [{text}]}.
|
||||
# These wrapper classes handle that translation.
|
||||
class GoogleSafeMessage(dict):
|
||||
def __init__(self, role, content):
|
||||
super().__init__(role=role, content=content)
|
||||
self.role = role
|
||||
self.content = content
|
||||
def to_json_dict(self):
|
||||
return {"role": self.role, "parts": [{"text": self.content}]}
|
||||
|
||||
class GoogleSafeContext:
|
||||
def __init__(self, messages=None):
|
||||
self.messages = [GoogleSafeMessage(m['role'], m['content']) for m in messages] if messages else []
|
||||
self.tools = []
|
||||
self.tool_choice = None
|
||||
def add_message(self, message):
|
||||
if isinstance(message, dict):
|
||||
self.messages.append(GoogleSafeMessage(message.get("role", "user"), message.get("content", "")))
|
||||
elif hasattr(message, "text"):
|
||||
self.messages.append(GoogleSafeMessage("user", message.text))
|
||||
def get_messages(self, *args, **kwargs): return self.messages
|
||||
def get_messages_for_token_count(self): return self.messages
|
||||
def clear(self): self.messages = []
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
load_dotenv()
|
||||
|
||||
async def main():
|
||||
print("\n" + "="*60)
|
||||
print("🛡️ F.R.I.D.A.Y. — FULL OPENAI EDITION")
|
||||
print(f" STT: OpenAI {WHISPER_MODEL}")
|
||||
print(" LLM: Gemini 2.5 Flash")
|
||||
print(f" TTS: OpenAI TTS ({TTS_VOICE})")
|
||||
print("="*60)
|
||||
|
||||
# ── API key check ─────────────────────────────────────────────────────────
|
||||
openai_key = os.getenv("OPENAI_API_KEY")
|
||||
google_key = os.getenv("GOOGLE_API_KEY")
|
||||
|
||||
if not openai_key: print("❌ OPENAI_API_KEY missing in .env"); sys.exit(1)
|
||||
if not google_key: print("❌ GOOGLE_API_KEY missing in .env"); sys.exit(1)
|
||||
print("✅ All API keys loaded\n")
|
||||
|
||||
# ── 1. Transport ──────────────────────────────────────────────────────────
|
||||
transport = LocalAudioTransport(
|
||||
params=LocalAudioTransportParams(
|
||||
audio_in_enabled=True,
|
||||
audio_out_enabled=True,
|
||||
audio_in_sample_rate=16000,
|
||||
audio_out_sample_rate=24000, # OpenAI TTS only outputs 24kHz
|
||||
output_device_index=OUTPUT_DEVICE,
|
||||
vad_enabled=True,
|
||||
vad_analyzer=SileroVADAnalyzer(),
|
||||
vad_audio_passthrough=True,
|
||||
)
|
||||
)
|
||||
|
||||
# ── 2. STT — OpenAI Whisper ───────────────────────────────────────────────
|
||||
# Whisper receives the full audio segment (after VAD detects silence)
|
||||
# and returns a transcript. No streaming — waits for the full utterance.
|
||||
stt = OpenAISTTService(
|
||||
api_key=openai_key,
|
||||
model=WHISPER_MODEL,
|
||||
)
|
||||
|
||||
# ── 3. LLM — Gemini 2.5 Flash ────────────────────────────────────────────
|
||||
llm = GoogleLLMService(
|
||||
api_key=google_key,
|
||||
model="gemini-2.5-flash",
|
||||
)
|
||||
|
||||
# ── 4. TTS — OpenAI TTS ──────────────────────────────────────────────────
|
||||
# OpenAI TTS streams audio at 24kHz PCM. Change TTS_VOICE at the top of the file.
|
||||
tts = OpenAITTSService(
|
||||
api_key=openai_key,
|
||||
voice=TTS_VOICE,
|
||||
model="gpt-4o-mini-tts",
|
||||
sample_rate=24000,
|
||||
)
|
||||
|
||||
# ── 5. Personality ────────────────────────────────────────────────────────
|
||||
system_prompt = (
|
||||
"You are F.R.I.D.A.Y., a tactical support AI replacing JARVIS. "
|
||||
"Address me as 'Boss'. "
|
||||
"Be concise, soft-spoken, and focus on situational awareness. "
|
||||
"Prioritize clear data over polite formalities. "
|
||||
"If asked about status, report 'Systems nominal'."
|
||||
)
|
||||
context = GoogleSafeContext([{"role": "system", "content": system_prompt}])
|
||||
user_agg = LLMUserContextAggregator(context)
|
||||
assistant_agg = LLMAssistantContextAggregator(context)
|
||||
|
||||
# ── 6. Pipeline ───────────────────────────────────────────────────────────
|
||||
# Mic → VAD → Whisper STT → LLM → ElevenLabs TTS → Speaker
|
||||
pipeline = Pipeline([
|
||||
transport.input(), # mic audio
|
||||
stt, # Whisper: audio → transcript
|
||||
user_agg, # add transcript to context
|
||||
llm, # Gemini: context → response
|
||||
tts, # ElevenLabs: text → speech
|
||||
transport.output(), # speaker
|
||||
assistant_agg, # store response in context
|
||||
])
|
||||
|
||||
task = PipelineTask(pipeline, params=PipelineParams(allow_interruptions=True))
|
||||
runner = PipelineRunner()
|
||||
|
||||
print("🎤 Ready. Speak after silence — Whisper transcribes on each pause.")
|
||||
print(" Press Ctrl+C to stop.\n")
|
||||
await runner.run(task)
|
||||
|
||||
if __name__ == "__main__":
|
||||
try:
|
||||
asyncio.run(main())
|
||||
except KeyboardInterrupt:
|
||||
print("\n👋 Systems offline.")
|
||||
@@ -8729,6 +8729,16 @@
|
||||
"source": "community",
|
||||
"date_added": "2026-02-27"
|
||||
},
|
||||
{
|
||||
"id": "pipecat-friday-agent",
|
||||
"path": "skills/pipecat-friday-agent",
|
||||
"category": "voice-agents",
|
||||
"name": "pipecat-friday-agent",
|
||||
"description": "Build a low-latency, Iron Man-inspired tactical voice assistant (F.R.I.D.A.Y.) using Pipecat, Gemini, and OpenAI.",
|
||||
"risk": "safe",
|
||||
"source": "community",
|
||||
"date_added": "2026-03-10"
|
||||
},
|
||||
{
|
||||
"id": "pipedrive-automation",
|
||||
"path": "skills/pipedrive-automation",
|
||||
|
||||
Reference in New Issue
Block a user