teyrebaz33 119bad65fc feat: prune old tool outputs before context compaction ...

Port the useful part of PR #588 onto current main without regressing
summary role alternation or the centralized call_llm-based summary path.

This adds a prune-first compression pass that:
- protects recent tool outputs with adaptive thresholds
- never prunes key tool outputs like read_file/memory/clarify
- skips the LLM summary call entirely when pruning alone is enough
- keeps head/tail protected windows untouched

Tests cover prune-only compaction, protected tools, and tail protection.

Co-authored-by: teyrebaz33 <hakanerten02@hotmail.com>

2026-03-13 21:21:28 -07:00

..

__init__.py

test: add unit tests for 8 modules (batch 2)

2026-02-26 13:54:20 +03:00

test_auxiliary_client.py

fix: vision auto-detection now falls back to custom/local endpoints

2026-03-09 15:36:19 -07:00

test_context_compressor.py

feat: prune old tool outputs before context compaction

2026-03-13 21:21:28 -07:00

test_model_metadata.py

test: comprehensive tests for model metadata + firecrawl config

2026-03-05 18:22:39 -08:00

test_prompt_builder.py

feat: secure skill env setup on load (core #688)

2026-03-13 03:14:04 -07:00

test_prompt_caching.py

test: add unit tests for 8 modules (batch 2)

2026-02-26 13:54:20 +03:00

test_redact.py

feat: secure skill env setup on load (core #688)

2026-03-13 03:14:04 -07:00

test_skill_commands.py

fix: improve gateway secret capture guidance message

2026-03-13 04:10:22 -07:00

test_subagent_progress.py

fix(display): fix subagent progress tree-view visual nits

2026-02-28 23:29:49 -08:00