What problem does it solve? Long-running agent sessions generate millions of tokens of conversation history that exceed context window limits, and naive compression loses critical details like file paths, error messages, and decision rationale, forcing costly re-exploration. ## Core Features & Use Cases - Anchored Iterative Summarization: Maintain persistent structured summaries with explicit sections for session intent, file modifications, decisions, and next steps, merging new content incrementally instead of regenerating from scratch. - Probe-Based Evaluation: Test compression quality with recall, artifact, continuation, and decision probes scored across six dimensions including accuracy and artifact trail. - Compression Method Selection: Choose between anchored iterative, opaque, and regenerative approaches based on session length, token savings targets, and re-fetching costs. - Use Case: A coding agent debugging a 401 authentication error across 178 messages compresses its history into a structured summary preserving the root cause, modified files, and failing tests, then validates the compression by probing whether the agent still knows which files it changed. ## Quick Start Ask the agent to compress the current conversation history into a structured summary with sections for session intent, files modified, decisions, and next steps.
