context-compression

Use anchored iterative summarization with explicit sections for files modified, files read, and decisions. Summarize only newly truncated content and merge it into existing sections rather than regenerating the full summary, which prevents cumulative detail loss across compression cycles.

Anchored iterative summarization scores highest (3.70 overall) for long sessions where file tracking matters. Opaque compression achieves the best ratio (99.3%) but sacrifices interpretability, while regenerative summaries suit sessions with clear phase boundaries.

Use probe-based evaluation instead of metrics like ROUGE. After compression, ask recall, artifact, continuation, and decision questions, then score responses across six dimensions including accuracy and artifact trail using an LLM judge.

Trigger at 70-80% context utilization for fixed thresholds, or use a sliding window keeping the last N turns plus a summary. Task-boundary triggers work when sessions have clear phase transitions like research then implementation.

Artifact trail is the weakest dimension across all compression methods, scoring only 2.2-2.5 out of 5. General summarization cannot reliably track files, so implement a separate artifact index or explicit file-state tracking outside the summarizer.

Never compress tool definitions, function call schemas, or API specs, since summarizing parameter names or types breaks the agent's ability to invoke tools. Also protect early turns containing task constraints and preserve code blocks and identifiers verbatim.