Foolllll 1daa0e3367 fix(compress): improve context compression, improve kv-cache rate of context compression, handle compression model modalities (#8530) ...

* fix(context): restore turn cap, serialize content parts and tool calls for llm compress, fix AftCompact debug log

Three context-compaction regression fixes after #8226:

1. Restore max_context_length -> enforce_max_turns propagation so
   normal turn-based truncation works again.
2. Serialize ContentPart and ToolCall objects into plain dicts in
   _message_to_dict so llm_compress no longer fails with JSON
   serialization errors.
3. Print _provider_messages (compacted) instead of run_context.messages
   (unchanged) in AftCompact debug log; truncate long role lists to
   first4,...,last4 to avoid log spam.

Assertions in tests are also hardened to avoid coupling to exact prompt
wording.

* fix(tool_loop_agent_runner): simplify context handling by removing redundant provider messages

* fix(tool_loop_agent_runner): rename context manager variables for clarity

* fix: update context compression to use recent token ratio instead of fixed count

* fix: enhance LLMSummaryCompressor to sanitize contexts and improve message handling

* ruff format

---------

Co-authored-by: Soulter <905617992@qq.com>

2026-06-03 17:40:05 +08:00

..

compressor.py

fix(compress): improve context compression, improve kv-cache rate of context compression, handle compression model modalities (#8530)

2026-06-03 17:40:05 +08:00

config.py

fix(compress): improve context compression, improve kv-cache rate of context compression, handle compression model modalities (#8530)

2026-06-03 17:40:05 +08:00

manager.py

fix(compress): improve context compression, improve kv-cache rate of context compression, handle compression model modalities (#8530)

2026-06-03 17:40:05 +08:00

round_utils.py

fix(compress): improve context compression, improve kv-cache rate of context compression, handle compression model modalities (#8530)

2026-06-03 17:40:05 +08:00

token_counter.py

feat: context token counting support for multimodal content (images, audio, and chain-of-thought) (#6596)

2026-03-19 17:25:49 +08:00

truncator.py

fix: prevent truncation logic from removing the only user message in long tool-calling conversations (#6198)

2026-03-20 12:05:54 +08:00