Teknium e75f58420c feat(compressor): major context compaction improvements ... Six improvements to reduce information loss during context compression, informed by analysis of Cline, OpenCode, Pi-mono, Codex, and ClawdBot: 1. Structured summary template — sections for Goal, Progress (Done/ In Progress/Blocked), Key Decisions, Relevant Files, Next Steps, and Critical Context. Forces the summarizer to preserve each category instead of writing a vague paragraph. 2. Iterative summary updates — on re-compression, the prompt says 'PRESERVE existing info, ADD new progress, UPDATE done/in-progress status.' Previous summary is stored and fed back to the summarizer so accumulated context survives across multiple compactions. 3. Token-budget tail protection — instead of fixed protect_last_n=4, walks backward keeping ~20K tokens of recent context. Adapts to message density: sessions with big tool results protect fewer messages, short exchanges protect more. Falls back to protect_last_n for small conversations. 4. Tool output pruning (pre-pass) — before the expensive LLM summary, replaces old tool result contents with a placeholder. This is free (no LLM call) and can save 30%+ of context by itself. 5. Scaled summary budget — instead of fixed 2500 tokens, allocates 20% of compressed content tokens (clamped to 2000-8000). A 50-turn conversation gets more summary space than a 10-turn one. 6. Richer summarizer input — tool calls now include arguments (up to 500 chars) and tool results keep up to 3000 chars (was 1500). The summarizer sees 'terminal(git status) → M src/config.py' instead of just '[Tool calls: terminal]'.		2026-03-21 08:14:14 -07:00
..
__init__.py	Refactor Terminal and AIAgent cleanup	2026-02-21 22:31:43 -08:00
anthropic_adapter.py	fix: crash on None entry in tool_calls list during Anthropic conversion (#2209)	2026-03-20 12:01:42 -07:00
auxiliary_client.py	fix: persistent event loop in _run_async prevents 'Event loop is closed' (#2190)	2026-03-20 09:44:50 -07:00
context_compressor.py	feat(compressor): major context compaction improvements	2026-03-21 08:14:14 -07:00
copilot_acp_client.py	fix(acp): preserve leading whitespace in streaming chunks	2026-03-20 09:38:13 -07:00
display.py	fix(display): suppress spinner animation in non-TTY environments	2026-03-20 12:52:21 -07:00
insights.py	fix(security): eliminate SQL string formatting in execute() calls	2026-03-19 15:16:35 +01:00
model_metadata.py	fix: add dashscope-intl.aliyuncs.com to URL-to-provider mapping	2026-03-20 12:51:39 -07:00
models_dev.py	fix: 6 bugs in model metadata, reasoning detection, and delegate tool	2026-03-20 08:52:37 -07:00
prompt_builder.py	feat: priority-based context file selection + CLAUDE.md support (#2301)	2026-03-21 06:26:20 -07:00
prompt_caching.py	fix(cache_control) treat empty text like None to avoid anthropic api cache_control error	2026-03-13 18:08:46 -07:00
redact.py	feat: secure skill env setup on load (core #688)	2026-03-13 03:14:04 -07:00
skill_commands.py	fix: disabled skills respected across banner, system prompt, slash commands, and skill_view (#1897)	2026-03-18 03:17:37 -07:00
smart_model_routing.py	feat: integrate GitHub Copilot providers across Hermes	2026-03-17 23:40:22 -07:00
title_generator.py	feat: auto-generate session titles after first exchange	2026-03-17 04:14:40 -07:00
trajectory.py	Refactor Terminal and AIAgent cleanup	2026-02-21 22:31:43 -08:00
usage_pricing.py	feat: use endpoint metadata for custom model context and pricing (#1906)	2026-03-18 03:04:07 -07:00