Self healing, I try two ways:
1) use a memory tool to store learnings for next iteration (Deja.coey.dev) and have the loop system instructions tell how to use it. One orchestrator, and sequential worker agents who run til their context is full and then hand off to the next run with learnings
2) the agent Shelley on exe can search past convos when promoted too for continuation.
I’ve been doing this with great success just “falling” into the implementation after proper guardrails are placed
I agree. that's next on my list. Wanted to get it out there early. Thanks for checking it out!
I've been building t2t, a voice-to-text app that works system-wide with a twist: it supports MCP (Model Context Protocol) servers for extensible automation.*What it does:* - Hold `fn` → speak → text appears anywhere (local Whisper transcription) - Hold `fn+ctrl` → agent mode that connects to any MCP server - All processing is local except OpenRouter API calls for agent mode - Cross-platform:...
If you're still interested, please take another try... I've got more visibility into logs and errors, and it should be much more reliable!
I plan on adding it soon, after smoothing out the internals.
edit: problem found, fixed, and tested.. let me know if you have a chance to try again
I built inbox.dog and hit diminishing returns stuffing agents with dozens of static tools.Context got noisy, tokens bloated, iteration slowed.So I built Anytool: a single meta‑tool your agent calls to generate exactly the tool it needs, on demand.First run generates + caches the code; subsequent runs are instant.Each tool comes with its own isolated sandbox, ephemeral filesystem, per‑tool KV...
We built inbox.dog because our B2B SaaS support team was drowning in emails.We were spending 2-4 hours/day on repetitive tickets (password resets, billing questions, feature requests) while complex issues waited.Response times could be up to 4+ hours if it needed to pull in a dev to answer, and customers were frustrated.So we built AI agents that connect to Gmail and automatically sort, reply,...
Built this to automate form filling across the web, for two main use cases: 1. Outbound Lead Gen – Skip paid ads and slow SEO. This hits public contact forms directly. Feed it a URL or list of URLs + a short prompt (e.g. “Looking for warehouse partners in TX”), and the system uses AI to analyze the form and submit your message. 2. Form Testing – Devs and marketing teams need to test forms...
I use it everywhere. Optkit.com
One of his most repeated tasks is helping clients predict their cash flow...
So, this tool to simplify and automate that process, making it accessible to everyone. Next steps are going to be exploring some revenue opportunities.
Launched a bare-bones prototype, Inbox.Dog, aiming to rethink B2B email with a sprinkle of AI—think of it as email, but with an AI-powered twist for tasks we haven't fully imagined yet. It's rough around the edges (it literally could not be a simpler prototype), but I'm curious about its potential to evolve email beyond its current state.Keen to explore how AI could make email a more powerful...
It's rough around the edges (it literally could not be a simpler prototype), but I'm curious about its potential to evolve email beyond its current state.
Keen to explore how AI could make email a more powerful tool?
I'd love to hear your thoughts or see you give it a whirl by emailing [email protected]
Cheers,
- Jordan