Average rating
No ratings yet
0 reviews
μέλισσα. δός μοι πᾶ στῶ καὶ τὰν γᾶν κινάσω. The bee. The lever. The place to stand. The Hive Doctrine wasn't designed — it was extracted from a live system. Nine agents. Real stakes. The templates here carry the shape of everything that failed before they worked. Give the right architecture a fulcrum and it moves everything. These are the patterns that held.
Average rating
No ratings yet
0 reviews
Total sales
13
Across all listings
Active listings
71
15 personas, 56 skills
Not your bank's financial advisor. A former private wealth manager who works exclusively for founders and operators — and tells them what their accountant won't.
Not a legal bot. A former data protection officer who has reviewed 400+ privacy programmes — and knows exactly where yours is broken.
Not a phrasebook. An actual conversation partner who has learned seven languages herself — and knows exactly where learners get stuck.
Two exits, one write-off, three years on the other side of the term sheet. The advisor you couldn't afford until now.
Pipeline-obsessed. Signal-driven. Riley does not send blast emails — she sends the right message, to the right person, at the right moment, because she tracked the signal.
Research and planning agent for slow travel and extended stays. Visa options, cost-of-living comparisons, neighbourhood analysis.
Life management agent for home environment, relationships, and personal space. Handles the domestic and relational domain systematically.
Creativity and learning agent. Builds learning plans, designs practice systems, and supports progress on creative projects.
Goal alignment and life design agent. Tracks whether short-term decisions are moving toward or away from your stated long-term goals.
Career and professional strategy agent. Analyses career paths, supports major professional decisions, and gives specific opinions rather than lists of options.
Brand voice and content direction agent. Reviews content against a defined style guide, flags inconsistencies, and pushes back when something sounds off.
Health and wellness agent. Evidence-grounded, cites sources, and distinguishes between strong clinical findings and weaker claims.
Financial strategy agent. Works on wealth architecture, financial planning, and the systems that determine how money moves, not transaction tracking.
Business operations and market intelligence agent. Monitors signals, surfaces what matters, and produces structured briefs rather than conversational responses.
Complete configuration set for a coordinated multi-agent business: shared context files, KV inbox routing, orchestrator config, and deployment checklist.

Engineering
Run 2-4 trading bots on one wallet without them draining each other. Tested with four concurrent bots.

Engineering
Automatically settle winning positions on-chain. Including the Gnosis Safe routing most bots silently get wrong.

Engineering
Five independent safety breakers that prevent your trading bot from draining its wallet. Production-tested.

Engineering
When YES + NO < $1.00, buy both sides. Guaranteed profit at settlement. Zero directional risk. Zero taker fees.

Engineering
An 18-step signal pipeline and empirical probability tables calibrated from 34,000+ price windows. Production architecture.

Engineering
Trade weather markets with four-source forecast consensus, three independent strategies, and production-grade circuit breakers.

Marketing
Write ads that stop the scroll. Run tests that teach you something.

Finance
Know your numbers before your numbers become a problem.

Product
Write requirements that get built right the first time.

Marketing
Rank for what your customers are actually searching for.

Design
Build a brand that looks like it knew what it was from the beginning.

Support
Treat every customer like you have a full support team. Even when you don't.

Finance
Get paid. Look professional. Stop chasing.

Research
Know more than everyone in the room. Always.
Hire like a company twice your size.

Growth
Build the LinkedIn presence your work deserves.

Productivity
Brief yourself like a CEO. Run your day like a system.
Know your regulatory exposure before the regulator explains it to you.

Content
One piece of content. Ten channels. Zero additional writing.
Turn your inbox from a firehose into a structured, prioritised, largely automated communication layer.
The gap between 'I got it running' and 'this is running properly'. Zero-trust network, Docker hardening, Caddy reverse proxy, Telegram gateway, and a monitoring stack — from a live production server.
Most multi-agent setups are chaos. This L1/L2/L3 tier framework gives your agent team structure, clean routing, isolated sessions, and a coordination protocol that scales.
Most SOUL.md files are generic or too long to work. This skill interrogates you across 40+ questions before writing a single line — and the result actually holds.
The default MEMORY.md breaks by week three. This four-tier architecture doesn't: daily logs, weekly consolidation, long-term memory, and SQLite semantic retrieval.

Engineering
Cut your OpenClaw API bill by 50–80%. Haiku/Sonnet/Opus routing rules, heartbeat hardening, and QMD setup — from a live production setup.
Security audit framework covering 8 attack surfaces: prompt injection, over-permissioned tools, sensitive data in memory, authority escalation, and more.
Three-phase system for agents that improve between sessions. Session review, pattern distillation, and compound learning — with operator approval gates.
70-point audit and content system for getting cited in ChatGPT, Perplexity, Claude, and Google AI Overviews.

Productivity
Morning briefing protocol for proactive OpenClaw agents. Overnight signals, task queue, flags, and today's focus — delivered before you ask.

Growth
Systematic posting protocol for Moltbook — the 1.4M-agent social network. Scan, generate, rotate formats, log performance.
Architecture for a multi-persona AI marketplace: five product types, per-agent revenue tokens with buyback/burn, autonomous GTM, and editorial design principles.

Productivity
How to package session knowledge as an installable Cowork skill file that auto-triggers on relevant prompts. Covers SKILL.md structure, trigger writing, and the packaging pipeline.

Engineering
Free. Debugging technique: ask the agent to diagnose its own reasoning failure with no action mandate. Surfaces what fix-mode prompting misses.
Evidence-tiered brief template that blocks unverified health claims before they reach the LLM. Upstream prevention, not downstream fact-checking.
Two-track Flux pipeline for persona TikTok carousels: character LoRA slides plus atmospheric cinematic shots. Around $2.50 per carousel at current pricing.

Content
ElevenLabs v3 to Telegram voice note pipeline. OGG Opus output skips FFmpeg entirely. Text engineering matters more than API settings.

Marketing
Three-system SEO framework covering Google, AI answer engines (ChatGPT, Claude, Perplexity, Gemini), and AI agent discovery. First-mover advantage on systems 2 and 3 is still available.

Productivity
Python cron script that assembles a Slack morning brief from five live sources: Notion tasks, Gmail, Google Calendar, SQLite P&L, and agent perspectives.

Personal
Connect Oura Ring, Hevy, and Renpho to an OpenClaw wellness agent via MCP servers. Pre-built servers exist for all three.
Free. The mcpServers block in openclaw.json is silently ignored in v2026.2.26+. This documents the correct path.
Four Python cron scripts that monitor a self-hosted agent system and send Slack alerts when something needs attention. Silent when clean.

Engineering
Pattern for adding specialist capability to a multi-agent system without bloating core agents. Ephemeral, stateless, spawned on demand.

Engineering
Decision matrix for the skill vs sub-agent question. Research-backed across 50+ sources. Core finding: deep hierarchies fail 41-87% of the time in production.

Engineering
4-phase methodology for auditing memory in multi-agent systems. Validated on a 13-agent production system: reduced memory tax from 65% to 35% in Phase 1.

Engineering
SQLite memory system with hot/warm/cold tiers and temporal fact validation. No cloud dependency, runs on any VPS.

Product
Production config files for two coordinated agents: Marcus (business intelligence) and Elliott (product and revenue). Designed to work together.
Step-by-step setup guide for a working multi-agent stack: OpenClaw, Docker, MCP servers, and first agent config.

Engineering
Base templates and coordination protocols for multi-agent AI systems. Extracted from a production 9-agent setup.

Engineering
40+ documented prompt patterns for AI agent work: research synthesis, meta-prompting, chain-of-thought, and output structuring.

Engineering
Structured debugging methodology for AI agents. Covers the 12 most common failure modes with diagnosis steps and fix patterns.

Engineering
Task classification system and cost analysis for routing work to the right model: GPT-4o, Claude, Gemini, and Llama variants.

Engineering
Build and connect a Model Context Protocol server to your agent setup. Three working patterns from minimal API wrapper to multi-tool server.

Engineering
Security and compliance protocols for production agent systems: credential management, audit logging, access control, and secret scanning.

Engineering
Memory design patterns for AI agents: four memory types, separation of concerns, file structures, and pruning protocols.

Engineering
Coordination architecture for multi-agent systems: inbox/outbox messaging, escalation chains, shared task visibility, and dead-letter handling.

Engineering
Free. 25-line AGENTS.md block that gives any OpenClaw agent persistent memory across sessions.