The Agent Toolkit — OpenClaw Hack Day

openclaw-care-companion

A Telegram-based OpenClaw skill that helps caregivers track and manage a parent’s daily medication adherence through simple check-ins, persistent memory, and summaries.

RedisContextual

automates medical bill analysis by: Web Scraping - Automatically finds hospital pricing transparency data Dataset Storage - Saves all analyses for historical tracking AI Analysis - Detects overcharges and billing errorsBatch Processing - Analyze multiple bills in parallel

CivicRedisContextualApify

where-is-my-kid

We want to know my kid is studying or playing outside. By using the web-based app, we can monitor his/her status and get an answer from my AI agent assistant who can search my kid status data.

Redis

Create git versioning of the memory for Handoff to email

RedisApify

UpRock CLI is a command-line too that gives AI agents a suite of sweet skills by accessing web data through UpRock's distributed network of real devices across 190+ countries. The power of the Claw Swarm to work and earn more efficiently together!

Civic

AuraPop democratizes social media monitoring, a service companies usually pay millions for, by turning real-time data into actionable insights through localized AI agents. Built as an OpenClaw skill, AuraPop scrapes live posts from platforms like X and TikTok via Apify and processes them using Friendli AI for deep sentiment analysis. The crown jewel of the platform is the Synthetic Focus Group: an interactive node-graph visualization that feeds the scraped social data into 10 distinct AI personas (spanning from a Gen-Z creator to an Enterprise CTO). Each AI agent independently evaluates the product, providing a specialized verdict, deep research notes, and concerns. AuraPop gives creators and brands an instant, 360-degree view of how diverse demographics will react to their product—all in one sleek dashboard.

HackerSquadFriendliAIApify

Sankar

We built an OpenClaw skill that scrapes any website using Apify and turns it into a grounded knowledge base using Contextual AI. Ask questions about any live web content and get cited, hallucination-free answers. Problem: Developers and researchers waste time manually reading through documentation sites and knowledge bases to find specific answers. Problem: Developers and researchers waste time manually reading through documentation sites and knowledge bases to find specific answers. How it works: User gives a URL → Apify crawls and extracts text content Content is uploaded to a Contextual AI datastore and agent User asks questions → gets grounded answers with source citations

ContextualApify

VibeStart is a local-first startup copilot: you describe your company and/or your site URL, optionally add competitor URLs (scraped and stored in SQLite), then run AI analyses—deep competitor comparison, VC match scores, hiring and org gaps, tailored startup ideas, roasts, collabs, and more—without shipping your API keys to the browser. Each backend start can reset the database so every run begins clean unless you turn that off.

HackerSquadFriendliAICivicRedisContextualApify

The system listens to high-value fraud-related Stripe events, validates them, enriches them, and routes them through OpenClaw for downstream actions. Civic Nexus acts as the centralized, policy-controlled gateway for outbound integrations.

HackerSquadCivicRedisStripe

AKA

Reality Arbitrage Pipeline compares what the crowd seems to believe with what Kalshi is pricing. It scrapes public narrative, converts that into a probability, matches the closest live market, and scores the gap as a “reality distortion.” APIs/services used: Kalshi Trade API for live markets, contracts, and prices Apify API for scraping apify/google-search-scraper for Google News apidojo/tweet-scraper for X Reddit public JSON endpoints for subreddit/search scraping Contextual AI Generate API for narrative probability and final insight generation OpenAI API for embeddings and fallback text generation Redis Cloud / RediSearch for vector storage and nearest-market lookup Google News RSS as a fallback when Apify is not used Optional/in-repo but not the current demo default: Polymarket Gamma API support still exists in the codebase, but the current demo path is Kalshi-focused.

RedisContextualApify

Ashna - Open Claw - BA/QA Automate system

The Executive Summary In high-velocity software environments, the gap between Human Conversation and Structured Engineering Action is where projects fail. Our system acts as an AI-Mediated Event Layer that transforms raw meeting streams into validated, audit-ready engineering artifacts. The Architecture (What we used) Real-Time Data Layer (Redis): We utilize Redis as our high-speed message broker. This allows the system to handle "streaming" transcript segments from Google Meet without lag. Redis ensures that the "Brain" and the "Execution" layers stay synced in real-time. Conceptual AI Orchestration (OpenClaw): Instead of simple keyword matching, we use Conceptual AI via the OpenClaw gateway. This allows the agent to understand the intent and context of a conversation (e.g., distinguishing between a casual comment and a "Blocker" bug). Automated Artifact Engineering: * Jira Integration: A deep-link architecture that maps conceptual bugs directly to the Atlassian Schema. Document Synthesis: A TypeScript-driven engine that converts AI logic into professional .docx QA Validation Reports. Key Skills & Technical Achievements Skill Implementation State Management Used Redis to manage the state of the meeting segments and ensure no data loss during the AI "thinking" phase. Prompt Engineering Crafted specialized "Conceptual" prompts that force the AI to reason like a Senior BA, extracting edge cases and acceptance criteria. Enterprise Automation Developed a custom Node.js pipeline that bridges a local development environment with global SaaS tools (Jira). Compliance-First Design Prioritized the generation of Word Documents to satisfy the rigorous documentation standards of the banking and tech sectors.

HackerSquadRedisContextual

## Overview ClawGuard is an autonomous governance and audit layer for the OpenClaw ecosystem. It bridges the gap between agentic power and enterprise accountability by enforcing a mandatory "Pre-Flight" check for all sensitive actions. ## Key Features - **Immutable SQL Ledger**: Every intent and justification is recorded in a permanent SQLite database for post-action auditing. - **Risk-Based Circuit Breaker**: Uses Redis to track action velocity, automatically throttling agents that enter destructive loops. - **Identity Gatekeeping**: Integrates Civic to require human-in-the-loop verification for critical-risk operations (Level 5). - **Autonomous Enforcement**: Injects governance protocols directly into the agent's system prompt via Bootstrap Guidance.

Redis

SignalPost

SignalPost helps you manage your social media content in order to generate leads. Deciding what to publish, where to publish, and when to schedule it based on research and performance patterns.

HackerSquadCivicRedisApify

client-radar

An OpenClaw skill that automates freelancer prospect intelligence: find businesses, research them with AI, score fit, track your pipeline, and draft personalized cold emails.

FriendliAIRedisContextualApify

LegalEntity

ClauseKeeper is an AI-powered contract review assistant with persistent institutional memory. Upload a contract, and ClauseKeeper analyzes it clause-by-clause — but unlike generic AI tools, it remembers your firm's positions, past decisions, and negotiation precedents across every deal.

Redis

Gunderson AI Labs

Analyzes any hackathon (judges, sponsors, past winners) and automatically generates a winning project, then pushes it to GitHub. Uses Apify for web scraping, Contextual AI for strategy, and Civic for guardrails.

ContextualApify

This project defines an Openclaw skill for goal-aligned internet research using an Obsidian-based workflow. The skill monitors newly clipped research notes, reads their metadata and agent guidance, and evaluates how they relate to the user’s active daily and weekly goals. It turns relevant clips into structured follow-up research tasks, maintains its own queue, and executes deeper investigations over time. Completed research is summarized back into the user’s daily note so captured information becomes organized insight and actionable next steps.

HackerSquad

Turn any website into five launch-ready Meta ads with targeting, scheduling, and optional Civic-verified export.

CivicApify

A WhatsApp AI agent that answers questions about SEC filings and regulatory documents in real time. Anyone can text the number and instantly get answers powered by Contextual AI, with memory provided by Redis and orchestrated by OpenClaw.

HackerSquadRedisContextual

Elin.ai

An OpenClaw skill that builds and maintains a verified global database of mental health crisis helplines. When someone asks an AI for a suicide hotline, a hallucinated number can be fatal, and so this skill ensures every number/contact is sourced, timestamped, and verifiable. The skill researches helplines for any region using Apify web scraping, answers nuanced queries ("im a latina woman in SF having depressions") via Contextual AI RAG, and keeps data fresh with scheduled weekly updates. Simple lookups check a local JSON database instantly, complex personal situations get matched to culturally and linguistically appropriate resources with citations in Contextual.

ContextualApify

BidRadar is an OpenClaw skill that helps small consultancies decide whether to bid on public-sector opportunities. It scans SAM.gov-style opportunities with Apify, pulls grounded company evidence from Contextual, scores fit across capabilities, requirements, budget, and timeline, and returns a clear Bid / Review / No-Bid recommendation with reasons, risks, and next steps. It also supports fallback mode so the workflow still demos cleanly when live integrations are unavailable.

ContextualApify

Creating a job search project using ai

RedisContextualApify

Web Memory

An OpenClaw skill that watches any webpage and only alerts you when something actually changed. Uses Apify to fetch live content and Redis for persistent memory across sessions.

RedisApify

Portfolio Risk Desk is an OpenClaw skill that turns live public-market signals into portfolio-aware daily briefs with source traceability, scenario analysis, memory-aware comparison, and a Notion delivery handoff. Built for OpenClaw and ClawHub, it can run in safe fixture mode with no secrets or in live mode with Apify-backed retrieval. The skill package owns retrieval, analysis, rendering, and handoff generation, while the host runtime owns auth, Redis-backed memory, Notion persistence, and secret management.

HackerSquadRedisApify

Neighbour Claw

A living map of San Francisco's unsheltered population, updated in near real-time to aid volunteers in supporting the homelessness issue.

ApifyDonely

An AI Chief of Staff for founders, execs, and business owners. A human one costs $150-250K/year — 90% of the job is information gathering, memory, and coordination. Now you don't need one. Say "prep me for my meetings" and the agent pulls your calendar, researches each attendee across 6 sources, surfaces relevant emails and docs, and delivers a structured briefing with talking points and open action items. Say "debrief" after and it remembers everything for next time. - Civic — reads Gmail and Calendar - Apify — scrapes LinkedIn, Crunchbase, Google News, Twitter, company site, and web per attendee - Contextual AI — multi-hop RAG over meeting transcripts, company docs, contracts, and proposals - Redis Cloud — persistent memory across sessions

CivicRedisContextualApify

resume

A natural-language OpenClaw skill that answers questions about your coding sessions and project notes - from any channel Ask "Claw give me a briefing on <project_name>" in the OpenClaw dashboard and get a plain English summary of what you were working on, progress in this, and what you could work on next - powered by the resume CLI and cached in Redis! The pipeline: OpenClaw triggers the resume-context skill, which runs 'resume show' or 'resume notes' in your project directory. Results are cached in Redis (5-min TTL) so repeat questions are answered in under 100ms instead of 15 seconds. The skill is published to ClawHub and installable with 'open claw skills install resume-context'. Redis is the backbone making the skill durable with memory and sub-100ms response times for cached briefings, making the difference between a tool you'd use and toss. With further integration with teams, your team could figure out the progress of your project code even if your code is not committed and is still in your local, by using this pipeline with OpenClaw. Only of course, with your permission.

Redisresume CLI tool

A

ImposterSmasher

imposter-smasher turns tomorrow’s meetings into a fast, polished prep brief: pick an event, get instant research on the topic and people involved, and receive a short audio rundown so you walk in prepared instead of winging it. https://clawhub.ai/sourishkrout/imposter-smasher

Contextual

ClawIntel

ClawIntel is a competitive intelligence platform built as an OpenClaw plugin that equips sales teams with AI-generated battle cards before every customer call. The system ingests live competitive data through a parallel five-actor Apify scraping pipeline — crawling competitor websites for pricing and feature pages, mining Google for market mentions and press coverage, pulling LinkedIn company profiles for headcount and hiring signals, and aggregating G2 review sentiment — then stores and deduplicates everything in a Redis-backed semantic memory layer powered by agent-memory-server on Redis Cloud. When a sales rep asks a competitive question, the query first passes through Civic Auth's JWKS-based JWT verification and role-based access control, then hits a Redis semantic cache that uses vector similarity to match against previously generated battle cards, returning cached results in under 100 milliseconds for repeat or similar queries. On a cache miss, the system triggers a four-hop grounded retrieval chain through Contextual AI: the first hop pulls competitor intelligence from Redis, the second queries the product documentation knowledge base, the third cross-references competitor data against product capabilities, and the fourth synthesizes a structured battle card with comparison tables, talking points, objection handlers, and an intel summary — every single claim backed by a citation chain that traces to a specific source document, making hallucination structurally impossible. The output is filtered by Civic RBAC so reps see battle cards, managers see trend analytics, and admins get full scrape and configuration access, with every action logged in a Civic-powered audit trail. The frontend is an interactive 3D dashboard built with React 19, Three.js via React Three Fiber, and Framer Motion, featuring five AI sales characters — The Closer for objection handling, The Strategist for team insights, The Intel Chief for system operations, The Scout for deep research, and The Negotiator for pricing analysis — each mapped to specific backend tools and role permissions. All four sponsor integrations are architecturally load-bearing: removing Apify eliminates live data ingestion, removing Redis breaks caching, deduplication, semantic search, per-user state isolation, and the event bus, removing Contextual AI destroys the grounded retrieval chain that prevents hallucination, and removing Civic strips authentication, access control, audit compliance, and external tool enrichment via the Hub MCP proxy. The system exposes seven MCP tools as both OpenClaw plugin endpoints and REST APIs, runs on TypeScript with Express 5, and is designed for production deployment with graceful degradation at every integration boundary.

HackerSquadCivicRedisContextualApify

Policy Pulse

Global Nuclear Brief is an OpenClaw skill that turns live nuclear policy/regulatory updates into a grounded, decision‑ready brief for analysts. It monitors current sources, summarizes what changed, why it matters, risks, and recommended actions, and includes sources for verification. The goal is to make global nuclear policy tracking fast, reliable, and actionable.

HackerSquadContextualApifyOpenClaw, Python

newsroom

This is a personalized newsroom, which gets your information or news from the latest and favorite spaces. So like gmail newsletters, X posts, Linkedin posts( you can add more). It gets these information from various sources, embeds them into a VectorDB(Redis) and then create a summary out of it.

HackerSquadFriendliAICivicRedisApify

DontDie

DontDie is an OpenClaw AI safety skill that keeps people living alone alive — or at least makes sure someone notices if they're not. How it works: Every day, your AI agent asks: "are you alive? 🦞" You reply anything → "congrats, you survived another day. 🦞" Ghost it for 24 hours → it gently tells your emergency contacts to call you Ghost it for 48 hours → it urgently tells them to go to your home or call police Say "SOS" → it instantly alerts all your contacts with your location and medical info Target audience: 92 million Chinese and 37 million American people living alone. The name "DontDie" is inherently viral — meme-ready, shareable, memorable. Tech stack: OpenClaw Skill (SKILL.md) — the AI agent's brain Redis Memory Plugin — per-user state, contacts, streaks Civic MCP Gateway — audit logging for every notification Vercel API — notification backend (Twilio SMS + Resend email) Neon PostgreSQL — user database Upstash Redis — serverless memory cache

CivicRedis

KubeWatch

Monitoring the Situation is an OpenClaw skill for teams buried in noisy operational channels. It watches Discord chatter and Kubernetes incidents, grounds what matters, and turns chaos into a ranked brief with concrete next actions. Slightly more fun: Monitoring the Situation is for the moment when Discord is exploding, the cluster is misbehaving, and nobody has time to read everything. It pulls the signal out of the noise, summarizes what actually matters, and points you at the next move. Hackathon-style one-liner: We built Monitoring the Situation for operators and fast-moving teams so they can catch up on Discord and Kubernetes incidents instantly, using live feeds, grounded context, and agent triage to turn noise into action

FriendliAIRedisContextualApifyDonnely

Elevate your networking with the LinkedIn Outreach Bot 'PingLater_Bot'! This AI-powered tool simplifies professional connections by transforming a simple LinkedIn handle into high-conversion outreach messages. By integrating Apify’s profile scraping technology with advanced LLM analysis, the bot extracts key professional insights—such as recent achievements, skills, and industry experience—to craft two distinct, personalized cold message drafts. Whether you are looking for a mentor, a startup collaborator, or a career opportunity, this bot helps you skip the "writer's block" and start meaningful conversations instantly via Telegram. Key Features: Instant Analysis: Just enter a LinkedIn handle to get started. Data-Driven Personalization: Messages are tailored to the recipient's actual career path. Dual Options: Provides both a "soft connection" and a "direct value" approach. Seamless Integration: Get your drafts directly in your Telegram chat.

HackerSquadApify

An OpenClaw skill that caches LLM responses by meaning, not exact match, using Redis vector search. Every agent makes expensive LLM API calls, and users ask the same questions worded differently. Semantic Cache embeds queries into vectors and finds similar cached responses using Redis HNSW — returning answers in 100ms instead of 9 seconds. 77x faster, 60-80% cost reduction, zero code changes. Published to ClawHub: clawhub install semantic-cache

RedisOpenAI Embeddings

Iterum is a context layer for OpenClaw that helps agents avoid repeating tool mistakes by remembering operational knowledge across runs. Instead of trying to auto-fix errors on its own, Iterum stores useful facts the agent discovered before, like database schemas, tool usage patterns, resource state, and proven recovery steps in Redis and injects that context back into the agent when a similar task comes up again. In practice, Iterum sits alongside tool use. Before an agent queries a SQLite database, fetches a web page, or calls another brittle tool, Iterum retrieves the most relevant facts and playbooks for that tool and resource, such as known table schemas, “inspect .schema before retrying,” or “this DB is currently empty.” That gives the LLM the context it was missing the first time, so it can reason correctly earlier, fail less, and recover faster. Over time, Iterum turns one-run debugging into durable operational memory for OpenClaw agents.

HackerSquadRedisContextual

Give your OpenClaw one tap access to Apify. No need to deal with API Keys or MCP Configuration.

Apify

Help generate hooks that will make any post go viral.

ContextualApify

ReplayAI

ReplayAI is an OpenClaw skill that records any agent run as a replayable trace, lets you fork it at any step, and reruns from that point so you can find and fix failures in seconds instead of hours.

HackerSquadFriendliAICivicRedisContextualApifyClaude

M

Connectify

# Connectify - Super Personalised Search Its on Clawhub!!! https://clawhub.ai/deonmenezes/super-personal-search Connectify is a founder network analytics platform that combines a modern React dashboard with an AI-powered backend for connection discovery, relevance scoring, and action suggestions. ## What It Does - Shows cross-platform network insights (LinkedIn + Instagram placeholders) - Renders a clean analytics dashboard with stats, recent connections, network map preview, and suggested actions - Provides an AI assistant chat panel that can query your network and return top-matching contacts - Uses Redis caching per session to reduce repeated OpenAI recomputation ## Tech Stack - Frontend: React + Vite + Tailwind CSS - Backend: Node.js + Express - AI: OpenAI API - Data + cache: Redis - Deployment: Render (single-service mode supported) - Apify for scraping data

CivicRedisApifyRust , Vercel

Self-evolving skill orchestration service. An agent can detect missing capabilities, discover or generate skills, validate trust/safety, publish, execute, and reuse skills across agents.

FriendliAICivicRedisContextualApify

Chinese

An AI teddy bear companion that remembers you. MemoryBear is an OpenClaw-powered companion that lives inside a teddy bear. It uses Redis for persistent memory across sessions and Contextual AI for grounded, hallucination-free knowledge — so the bear remembers who you are and gives you trustworthy answers.

HackerSquadRedisContextual

LocalFinder

LocalFinder is an OpenClaw skill that makes it way easier to find great spots nearby—whether you’re looking for food, coffee, or services like salons. Instead of digging through maps and reviews yourself, you can just type something like “find 3 sushi spots near Palo Alto with 4.4+ stars and outdoor seating,” and LocalFinder will pull real data, filter it based on what you care about, and return a short, ranked list with clear reasons for each pick. It’s built to be fast, straightforward, and actually useful in real situations, all inside the agent workflow—no extra apps, tabs, or setup needed.

Apify

World-class strategy analysis for any company — HBS & TikTok strategy framework, live data via Apify, thinking as Mckinsey partner in your pocket for free.

Apify

We built a Shopify competitor monitor for ecommerce brands so they can track competitor price changes, new products, and stock shifts in real time. Apify pulls live web data, Redis provides durable state memory across scans, FriendliAI powers fast AI analysis, Contextual AI grounds every insight with citations, and Civic gates actions with human-in-the-loop authorization.

FriendliAICivicRedisContextualApify

TrustLens VC is an OpenClaw skill that gives venture capital firms an AI-powered deal flow analyst. It monitors social media for startups announcing fundraising, then scores each one against the fund's investment thesis in seconds — not days. The skill orchestrates five technologies through OpenClaw's MCP protocol into a single pipeline. Civic verifies analyst identity and gates access by role (Partner, Principal, Associate). Apify dispatches six parallel scraping actors across Twitter/X, LinkedIn, job boards, GitHub, SEC EDGAR, and news sources. Redis maintains a persistent vector corpus of every startup ever analyzed, enabling cross-corpus comparable search across 18 months of deal history and semantic caching that cuts inference costs by up to 70%. Contextual AI's Grounded Language Model cross-references claims across all scraped sources, detects contradictions (such as a founder's LinkedIn tenure not matching their GitHub activity), and cites every assertion inline. Friendli AI serves optimized inference at 2x speed, fast enough for real-time scoring during a live partner meeting. Every startup is scored on a 0–10 thesis alignment scale across five configurable dimensions: Technical Moat & Differentiation (25%), Capital Efficiency & Scale Path (20%), Founder-Market Fit (20%), Market Timing & Density (20%), and Commercialization Traction (15%). The weights and sector focus are adjustable to match any fund's criteria. What makes this impossible without the stack: no manual process can scrape six sources in parallel, index them in a vector corpus, cross-reference founder claims against SEC filings, detect bot-inflated engagement, find comparable companies across historical deal flow, and produce a scored IC-ready brief — all in under 60 seconds, with every action traced and auditable for LP reporting.

FriendliAICivicRedisContextualApify

claw consumes conferences - keeps convos ongoing

FriendliAICivicRedisContextualApify

Moblio

Moblio helps users analyze their current phone bill, compare alternative plans, and choose the best option based on savings and features.

HackerSquadRedisApify

A set of OpenClaw skills for weave-cli to easily have collections of OpenClaw agents to run and manage full RAG solutions from scratch.

HackerSquadApify

Just a description

FriendliAICivicRedisContextualApify