The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
asciinema for AI coding agents.
Your AI coding agent just made 300 tool calls and changed 14 files. The diff shows what changed. It doesn't show what it read, what failed, or where it looped.
Claude Code and Codex already write a full log of every session to your home folder. Nobody opens it. So I built the viewer.
agentcast turns those logs into one self-contained HTML replay: a timeline, every edit as a diff, every command with its output, errors in red, tokens and estimated cost, and a "blast radius" list of every file the agent touched. Deep links like replay.html#142 so you can point a reviewer at the exact step.
Pure Python, zero dependencies, nothing leaves your machine. Secrets scrubbed by default.
Live demo: greenaisolution.github.io/agentcast
Code (MIT): github.com/GreenAiSolution/agentcast
If you ship agent work for clients, this is the receipt.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Trending
retrofuturism
Share your favorite designs from the past exploring the distant future. Or your own take on the aesthetic 🚀
halftone
Dots. Texture. Grain. Let's see your designs using the halftone technique.
Midjourney
Creatives are showing off what they're making with Midjourney.
colorgrade
Video editors sound off: How do you enhance the mood of your video in color grading?
Related posts
New here. My visibility score is 59 and I want to understand how this actually works.
If you've built up a profile on Contra what moved the needle? Projects, services, posting consistently, something else? I'd rather learn from people who've done it than guess.
Some context on where I'm coming from. I spent the last year on Upwork and hit Top Rated Plus with just over $100k earned. But I remember the start clearly: empty profile, no clients, no reviews, applying into silence. That part was not fun and it took a while.
So I know what this phase looks like. Empty profile again, score at 59, starting over. The difference is I've done it once, and I know it comes down to staying consistent when nothing's happening yet.
Here's my plan, and I'm posting it publicly so I actually stick to it: I'll ship work, publish every project, share what I'm building, and post updates as the profile grows including the parts that don't work.
I build AI agent systems and automation. Recent work: a multi-agent platform with 12 role-scoped specialists, a grant intelligence tool for nonprofits, and an n8n pipeline running unattended for a healthcare SaaS.
Follow along if you want to watch someone build a profile from zero. And if you've got advice on the discovery score, I'm listening.
Basically, when you hit certain milestones, the number goes up. Profile quality, number of likes, payment goals (1k, 5k, 10k...), client reviews... The higher the number, the better your visibility on the platform
AI Chatbots Should Do More Than Just Answer Questions 🤖
I’ve been building AI chatbot solutions that go beyond basic Q&A.
A useful business chatbot should be able to understand your company’s knowledge, answer customer questions accurately, capture and qualify leads, trigger workflows, and connect with the tools your team already uses.
Recent solutions I’ve worked on include:
→ Custom knowledge-base / RAG chatbots
→ OpenAI & LLM integrations
→ Website AI assistants
→ WhatsApp & messaging integrations
→ Lead capture and qualification
→ CRM & API integrations
→ Human handoff workflows
→ Conversation history & analytics
→ Custom admin dashboards
The goal is simple: turn AI into a practical business tool—not just another chat window.
If your business has repetitive customer questions, manual lead handling, or information scattered across documents and systems, a custom AI assistant can automate a significant part of that workflow.
This is wonderful. I fully buy into this content any day any time. Keep up the good work
Filters didn't hold. Labels did.
I spent last week trying to rob my own agents. Every filter I wrote to catch a malicious instruction eventually lost - the space of wordings is infinite. What held was provenance: label every value with where it came from, and check the label where the action happens, not the string where the text arrives. An agent can read a stranger's message. It just can't let that message pick the recipient.
The same shape holds between a model and a motor. Checking the destination is not enough - you have to check the swept path.
Both benches run offline, deterministic, and ship with their baselines. A zero with no baseline is a marketing number.
If you build agents that touch money or hardware, I would like to connect. I would rather be argued with than agreed with, and I will show you the code first. Both write-ups are on my profile.
