Why your agent keeps resurrecting blockers from tasks fixed weeks ago
Lopsided_Manner_7602 · reddit · 2026-09-22
The author walks through a real case: an agent left a working-memory note three weeks ago ("blocked on queue-name decision"), the task was finished by a human, yet the agent now brings it up on unrelated work.
Root cause: it's not hallucination — the agent stored a moment-in-time task state with the same weight and format as permanent knowledge ("this codebase uses PostgreSQL"). Nothing distinguished knowledge from working memory.
The fix isn't storing less, but sorting into two piles:
- Task-specific facts (blockers, half-done work, temporary decisions) live in a "for now" pile cleared when the task closes.
- A fact earns promotion to the "keep for good" pile only after it recurs across multiple tasks, via an explicit decision — never automatically.
Takeaway: an agent that remembers everything isn't smarter; it must learn to separate "still true" from "just happened," or stale state resurfaces with the authority of architecture decisions.
More from coding & agent
- Stripe's Emily Sands on agentic commerce: compute is the new cash, fraud is now full-funnel — every · 2026-09-22
- LangChain offers open-source decision model SemIf free for a week in LangSmith Gateway — hwchase17 · 2026-09-22
- Writing skills for models and their harnesses is one of the highest-leverage skills now — pvncher · 2026-09-22
- Grok VS Code extension now listed on Open VSX and the VS Code Marketplace — PawelHuryn · 2026-09-22
- LLMs Are Too Big: Aronchick on Right-Sizing Models for Log Routing — aronchick · 2026-09-22
- Dev syntax-highlights nested string literals inside a coding agent, making prompts readable again — lucasmeijer · 2026-09-22