Ezra 2da4a3d937 fix(context_compressor): reduce default protect_last_n from 20 to 5 ...

The previous default of 20 protected messages was overly conservative,
preventing meaningful compression on long sessions. Reducing to 5
strikes a better balance between preserving recent conversational
context and allowing the compressor to actually reduce token pressure.

Updates both ContextCompressor default and AIAgent integration,
plus adds a regression test verifying the last 5 turns are never
summarized away.

2026-04-06 17:24:39 +00:00

..

__init__.py

Refactor Terminal and AIAgent cleanup

2026-02-21 22:31:43 -08:00

anthropic_adapter.py

fix: preserve Anthropic thinking block signatures across tool-use turns

2026-04-02 10:30:32 -07:00

auxiliary_client.py

fix: use main provider model for auxiliary tasks on non-aggregator providers (#5091)

2026-04-04 12:07:43 -07:00

builtin_memory_provider.py

feat(memory): pluggable memory provider interface with profile isolation, review fixes, and honcho CLI restoration (#4623)

2026-04-02 15:33:51 -07:00

claw_runtime.py

[EPIC-999] Phase II — The Forge: claw_runtime scaffold + forge pipeline

2026-04-05 23:32:53 +00:00

context_compressor.py

fix(context_compressor): reduce default protect_last_n from 20 to 5

2026-04-06 17:24:39 +00:00

context_references.py

fix(security): add .config/gh to read protection for @file references (#4327)

2026-03-31 12:48:30 -07:00

copilot_acp_client.py

fix: bridge tool-calls in copilot-acp adapter

2026-04-06 01:47:57 -07:00

credential_pool.py

feat: centralized logging, instrumentation, hermes logs CLI, gateway noise fix (#5430)

2026-04-06 00:08:20 -07:00

display.py

feat: add inline diff previews for write actions

2026-04-01 02:13:57 -07:00

insights.py

fix(insights): show cache tokens in overview so total adds up (#4428)

2026-04-01 03:06:47 -07:00

memory_manager.py

fix: mem0 API v2 compat, prefetch context fencing, secret redaction (#5423)

2026-04-05 22:43:33 -07:00

memory_provider.py

feat(memory): pluggable memory provider interface with profile isolation, review fixes, and honcho CLI restoration (#4623)

2026-04-02 15:33:51 -07:00

model_metadata.py

fix(opencode-go): strip trailing /v1 from base URL for Anthropic models (#4918)

2026-04-03 18:47:51 -07:00

models_dev.py

feat: /model command — models.dev primary database + --provider flag (#5181)

2026-04-05 01:04:44 -07:00

prompt_builder.py

feat: add GPT/Codex execution discipline guidance for tool persistence (#5414)

2026-04-05 21:51:07 -07:00

prompt_caching.py

fix(prompt-caching): skip top-level cache_control on role:tool for OpenRouter

2026-03-21 16:54:43 -07:00

redact.py

fix: mem0 API v2 compat, prefetch context fencing, secret redaction (#5423)

2026-04-05 22:43:33 -07:00

skill_commands.py

fix(gateway): resolve Telegram's underscored /commands to skill/plugin keys

2026-04-05 11:59:28 -07:00

skill_utils.py

fix: respect per-platform disabled skills in Telegram menu and gateway dispatch (#4799)

2026-04-03 10:10:53 -07:00

smart_model_routing.py

Merge branch 'main' into rewbs/tool-use-charge-to-subscription

2026-04-02 11:00:35 +11:00

subdirectory_hints.py

feat: progressive subdirectory hint discovery (#5291)

2026-04-05 12:33:47 -07:00

title_generator.py

feat(agent): configurable timeouts for auxiliary LLM calls via config.yaml (#3597)

2026-03-28 14:35:28 -07:00

trajectory.py

Refactor Terminal and AIAgent cleanup

2026-02-21 22:31:43 -08:00

usage_pricing.py

fix: status bar shows 26K instead of 260K for token counts with trailing zeros (#3024)

2026-03-25 12:45:58 -07:00