My AI agents were choking on their own markdown logs. I gave them a database.

Description and use

AI agents lost context due to large markdown logs. PentaCore, a Rust‑based SQLite server, stores structured notes and entities, enabling quick context restoration, avoiding repeated failures, and sharing decisions among subagents.

Problem

AI agents accumulated markdown logs that exceeded the 256k token context window, causing restarts, loss of progress, repeated failures, and lack of shared state between subagents.

Outcome

Agents no longer choke on logs; context is restored with minimal tokens, failures are avoided, subagents receive decisions, and overall progress is summarized in a single query.

Workflow

• 1. Capture agent actions and results as structured notes and entities in PentaCore.
• 2. Store failures as marked checks to prevent repeats.
• 3. Propagate top-level decisions to subagents via shared entity status.
• 4. Use single queries (e.g., query_entities) to retrieve pending tasks or overall status.
• 5. On session start, call resume(task) to restore minimal context needed for continuation.

Evidence

Agents produce logs at every step: what they found, what they tried, what failed. All of that was landing in markdown files.

Source

I did not find a tool that fit. So I built one: PentaCore, an MCP server in Rust. One binary, one encrypted SQLite file.

Source

Now "what has not been checked yet" is one call: {"name":"query_entities",...}

Source