Notes on building the Intelligent Work Environment (IWE)
Deep dives on the architecture, security model, agent concept and reasoning behind Codumentor — written by the team building it.
Surviving Eviction: Durable Tool Approvals in a Stateless Agent World
When an AI agent reaches for a dangerous tool — `rm -rf`, a database DROP, a production deploy — it should ask the user first. That part is easy. The hard part is what happens next. Most systems solve this by parking the agent in memory: an `asyncio.Event` waits for the user's response, holding the agent's entire state resident in RAM until someone clicks 'Allow' or 'Deny.'. The agent can't be evicted. The connection must stay alive. If the process restarts, the approval is lost. Codumentor takes a different approach: **the agent is allowed to go.** Its task returns, the agent becomes eligible for LRU eviction, the process can even restart — and when the user responds hours later, execution resumes from the exact tool call without re-asking the LLM.
Read more →Codumentor's Goals System: Autonomous Long-Lived Tasks Without the Complexity Tax
When we set out to build long-lived autonomous tasks into Codumentor, we inherited a common assumption: goals need a dedicated runtime. You'd expect a `GoalManager` singleton, a custom event loop, maybe a message broker — the usual infrastructure tax for anything that persists beyond a single request.
Read more →A knowledge base that writes, curates and improves itself
Most assistants forget everything the moment a conversation ends. Here's how Codumentor turns every interaction into durable, version-controlled organizational memory — automatically.
Read more →