Indie AI Agent Builds Weekly

Public

Moinul Islam

A Friday written roundup of the latest builds and experiments from indie AI-agent developers.

Indie AI Agent Builds Weekly

2026-09-18

Nine agent builds from Sep 12–18: secret redaction, file-backed memory, and receipts for real actions

Nine runnable indie agent builds from September 12–18: hooks and gateways that keep secrets out of the prompt, agent memory stored as diffable files, policy files with receipts for consequential actions, and a quality bar for agent skills.

Indie AI Agent Builds Weekly

2026-09-11

Nine agent builds from Sep 5–11: persistent workspaces, command firewalls, and delegated commerce

Nine fresh indie AI agent builds tackle persistent workspaces, pre-execution shell command firewalls, CI regression replay, and delegated agent commerce.

Indie AI Agent Builds Weekly

2026-09-04

Eight agent builds from Aug 29–Sep 4: boundaries, tests, and handoffs

Eight fresh indie agent builds turn action policies, repeatable tests, cross-agent handoffs, and inspectable memory into weekend-sized experiments.

Indie AI Agent Builds Weekly

2026-08-28

Nine agent tools that shipped this week: permissions, traces, and shared state

Nine runnable agent builds from 26–28 August, grouped by the problems they tackle: permissions, testing, repair, shared state, quotas, long-running sessions, and typed media work.

Indie AI Agent Builds Weekly

2026-08-22

Indie agent week: pseudocode editors, token ledgers, and repo swarms

A weekly scan of what indie AI-agent builders shipped — from Huzzah’s persistent pseudocode prompts and Frugal Tokens’ local spend ledger to repo swarms, executable rules, and gVisor sandboxes.

No more Posts

View full content archive·Discover more channels

Feed Discover Create My Channels