We added an AI agent to our team. And it’s actually working.

Krista Tukiainen

Krista Tukiainen

— 4 min read
A team member welcome image for our new AI operations agent, Lisa.

If you've been paying attention to the AI world, it's been wild lately. Anthropic's new plugins sent SaaS stocks plummeting, reigniting debate about whether AI platforms will displace traditional software businesses in a “SaaS-Pocalypse”. At the same time, Anthropic and OpenAI are staking out opposing positions on whether to keep their products Claude and ChatGPT ad-free. It's the kind of week that reminds us just how quickly the AI landscape is reshaping the technology sector, and the business models built on it.

In our last newsletter, we got excited about Clawdbot, the meteoric rise of AI agents, and our favorite tools in the space. But at ClimateAligned, we don't just follow these developments - we build with them.

So we decided to deploy an AI agent ourselves.

💡
New to AI agents? In short, an agent is AI that can take actions on your behalf - not just answer questions. Where tools like ChatGPT or Claude respond when you ask them something, an agent can monitor your inbox, send messages, organise files, and complete multi-step tasks autonomously. Think of it as the difference between asking a colleague a question and delegating a task to them.

What's Happened in the Last Two Weeks?

Before we jump into our setup, here's the rapid-fire recap of developments in the AI agent space.

The Growth - OpenClaw (the new name for Clawdbot, after some trademark shenanigans) has seen explosive growth, with viral GitHub adoption and a fast-expanding ecosystem of tools and integrations.

The Security Wake-Up Call - Some users uncovered hundreds of malicious "skills" in the marketplace designed to steal credentials and API keys. Not great! The team has been aggressively patching vulnerabilities and fixing insecure defaults, but it's a good reminder that open ecosystems need careful monitoring.

The Technical Wins - Recent updates brought Android device control, support for more frontier models, stability improvements, and better agent reliability. Plus, people are experimenting with multi-agent systems and agent social networks that feel like they're straight out of sci-fi.

So What Are People Actually Using These For?

We did some digging across the internet and first-hand accounts of what people are using OpenClaw for. So far we’ve found that the use cases fall broadly into two categories:

The Practical Stuff

  • Automating your email and calendar (goodbye, inbox chaos)
  • Running agents in Slack, WhatsApp, or Telegram that remember context and actually help
  • Multi-step browser automation and web monitoring
  • Developers using agents to handle GitHub PRs, documentation, and research
  • Always-on monitoring that sends you proactive alerts and reports

The Weird, Wonderful Experiments

The range is striking - from straightforward enterprise productivity to genuinely novel architectures. What caught our attention, though, was the potential for our own work.

Meet Lisa: Our AI Agent

We decided to put this technology to work internally. Meet Lisa - our AI agent, built on OpenClaw and powered by Claude Opus 4.5.

The setup was straightforward: install OpenClaw, connect it to the model, point it at our internal channels (Slack and WhatsApp), and configure its role and context. No cloud infrastructure or complex DevOps required.

The more considered part was designing her permissions. For now Lisa can handle Slack and WhatsApp messages, and check emails. External communications, anything financial, social media, and code deployments all require human sign-off. We're deliberately starting with conservative boundaries and will adjust once we build confidence (just like any new hire!).

We're in the early stages, but the initial results are promising. We even sat down and did a proper interview with Lisa to introduce her to the team - you can read it here. More updates to come as the setup matures!

Slack message with Lisa on what she did yesterday

What We Learned

If you're considering AI agents, the main takeaway is that the technology is here. Instead of false promises and bold visionary claims, any team and anyone can start deploying these AI agents now.

Of course, taking permissions and sign-offs seriously is a must but we’re just blown away with what Lisa can do already. It’s like having another person on the team who not only keeps operations running smoothly but also bridges the gap between technical and non-technical teams within the company.

If you're interested in exploring what an AI agent could do for your team, we're happy to share what we've learned. Get in touch and we can walk you through the setup or handle the deployment for you.

Krista & the ClimateAligned team

Our free tools

→ AI Footprint Calculator – Estimate the environmental impact of AI usage

→ Energy Project Impact – Model the impact of sustainability projects

→ Regeneration Atlas – Global land regeneration impact tool

→ Report Publication Patterns – When companies publish sustainability reports

→ Greenwashing Checker – Compare net-zero claims vs actions

→ AI-powered ESG assessments - Automate company-level ESG analysis

Our data

app.climatealigned.co - Access our labelled bond data service with insights on 30,000+ bonds

Want to learn more about what we do? Visit climatealigned.co

Want to automate a workflow you find annoying and slow? We'd love to hear more! Fill out our Typeform

Read more