This product was not featured by Product Hunt yet. It will not be visible on their landing page and won't be ranked (cannot win product of the day regardless of upvotes).
Agent Desk + Session Intelligence
Find past AI coding sessions and learn what went wrong
Find past AI coding work and learn what to change next. Agent Desk brings sessions from Claude Code, Cursor, Codex, Copilot, Kiro and Antigravity into one local macOS app, with rules, skills and hooks in a separate harness view. Session Intelligence analyzes extracted transcripts for missing context, repeated corrections, outcomes and wasted effort. Two separate MIT-licensed tools. Build Agent Desk from source; run analysis with your own LLM.
The engineering teams getting real gains from AI have built a system around their coding agents. Across our conversations with engineering leaders this year, the same 5 building blocks kept showing up (last image in the gallery). We built many of these pieces and ran our own engineering on them first. Now we're open-sourcing them, one batch at a time.
Today's batch is two tools that work together.
Agent Desk, a Mac menu bar app that shows the prompts and harness behind your AI coding sessions - https://github.com/tetriz-ai/agent-desk - Your sessions in one view, project by project, across Claude Code, Cursor, GitHub Copilot, Antigravity, Codex and Kiro - The skills, commands, rules, hooks, subagents and instruction files shaping your agents, and whether each is committed, local or user-level - Runs locally. The open-source code has no upload functionality.
Session Intelligence, a Go library that evaluates coding sessions turn by turn - https://github.com/tetriz-ai/session-intelligence - Shows where your instructions worked and where the conversation went off track - Lets you compare prompting approaches and guide changes to your prompts or agent setup - Bring your own LLM
Start by browsing your sessions in Agent Desk. Use its companion extractor to export a transcript, then run Session Intelligence separately with your own LLM to understand where the session went off track and what to change next.
Agent Desk ships as source for macOS 12+. Apple silicon is tested; Intel Macs are unverified. Session Intelligence runs separately with your own LLM. The bundled Gemini integration requires your API key; using it sends transcripts to Gemini and incurs usage charges.
None of this is a cure-all. Every team has to fit it to its own context. Clone both, try them on a project your team is already working on, and tell us which block you'd want next.
What would you most want to understand about a coding session that went sideways?
Hey, congrats on the Agent Desk + Session Intelligence launch! We're sharing a launch day with Benson: https://www.producthunt.com/products/benson-2. Would appreciate a comment or a little support if you have time today. Wishing you a great launch!
About Agent Desk + Session Intelligence on Product Hunt
“Find past AI coding sessions and learn what went wrong”
Agent Desk + Session Intelligence was submitted on Product Hunt and earned 16 upvotes and 2 comments, placing #47 on the daily leaderboard. Find past AI coding work and learn what to change next. Agent Desk brings sessions from Claude Code, Cursor, Codex, Copilot, Kiro and Antigravity into one local macOS app, with rules, skills and hooks in a separate harness view. Session Intelligence analyzes extracted transcripts for missing context, repeated corrections, outcomes and wasted effort. Two separate MIT-licensed tools. Build Agent Desk from source; run analysis with your own LLM.
Agent Desk + Session Intelligence was featured in Open Source (68.9k followers), Developer Tools (521.2k followers) and Artificial Intelligence (481.2k followers) on Product Hunt. Together, these topics include over 234.9k products, making this a competitive space to launch in.
Who hunted Agent Desk + Session Intelligence?
Agent Desk + Session Intelligence was hunted by Prasanna R. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how Agent Desk + Session Intelligence stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
Prasanna R
Hi Product Hunter,
The engineering teams getting real gains from AI have built a system around their coding agents. Across our conversations with engineering leaders this year, the same 5 building blocks kept showing up (last image in the gallery). We built many of these pieces and ran our own engineering on them first. Now we're open-sourcing them, one batch at a time.
Today's batch is two tools that work together.
Agent Desk, a Mac menu bar app that shows the prompts and harness behind your AI coding sessions
- https://github.com/tetriz-ai/agent-desk
- Your sessions in one view, project by project, across Claude Code, Cursor, GitHub Copilot, Antigravity, Codex and Kiro
- The skills, commands, rules, hooks, subagents and instruction files shaping your agents, and whether each is committed, local or user-level
- Runs locally. The open-source code has no upload functionality.
Session Intelligence, a Go library that evaluates coding sessions turn by turn
- https://github.com/tetriz-ai/session-intelligence
- Shows where your instructions worked and where the conversation went off track
- Lets you compare prompting approaches and guide changes to your prompts or agent setup
- Bring your own LLM
Start by browsing your sessions in Agent Desk. Use its companion extractor to export a transcript, then run Session Intelligence separately with your own LLM to understand where the session went off track and what to change next.
Agent Desk ships as source for macOS 12+. Apple silicon is tested; Intel Macs are unverified. Session Intelligence runs separately with your own LLM. The bundled Gemini integration requires your API key; using it sends transcripts to Gemini and incurs usage charges.
None of this is a cure-all. Every team has to fit it to its own context. Clone both, try them on a project your team is already working on, and tell us which block you'd want next.
What would you most want to understand about a coding session that went sideways?