micro-models-agent 0.58.2 → 0.59.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +8 -0
- package/dist/i18n/en.json +2 -0
- package/dist/i18n/ru.json +2 -0
- package/dist/main.js +742 -389
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -4,6 +4,14 @@ All notable changes to Micro Models Agent (MMA) will be documented in this file.
|
|
|
4
4
|
|
|
5
5
|
The format is based on [Keep a Changelog](https://keepachangelog.com/), and this project adheres to [Semantic Versioning](https://semver.org/).
|
|
6
6
|
|
|
7
|
+
## [0.59.0] - 2026-09-05
|
|
8
|
+
|
|
9
|
+
### Added
|
|
10
|
+
- **Instructions overflow handling**: AGENTS.md and the project map that exceed the system-prompt budget are no longer silently dropped — they are summarized by the LLM into the remaining budget (disk-cached in `<project>/.mma/cache/prompt-summaries`, invalidated by file edits/budget changes) or truncated with a marker; a small essential hint block tells the model what was compressed and to `read_file` the full source
|
|
11
|
+
- **Startup overflow warning**: bootstrap dry-runs the system budget with the real block set and warns the user when instructions/project map will not fit, with a concrete fix — `mma context <N>` for a global `contextWindow`, or an edit of `provider.entries[].contextWindow` in `config/provider.json` when the active entry overrides it
|
|
12
|
+
- **Context probe**: at startup MMA asks LM Studio's native API for the context length the model is actually loaded with and compares it with the configured `contextWindow` — warning on overflow risk (no auto-clamp, user decides), info suggestion when the model supports more; result logged as a `context_probe` session entry
|
|
13
|
+
- **Config**: `instructions.summarize` (default `true`) to disable LLM summarization (truncation fallback only)
|
|
14
|
+
|
|
7
15
|
## [0.58.2] - 2026-09-01
|
|
8
16
|
|
|
9
17
|
### Fixed
|
package/dist/i18n/en.json
CHANGED
|
@@ -464,6 +464,8 @@
|
|
|
464
464
|
"repl.agents_label": "Instructions:",
|
|
465
465
|
"repl.not_found": "not found",
|
|
466
466
|
"repl.disabled": "disabled (--no-agents-md)",
|
|
467
|
+
"repl.context_probe_small": "Context probe: model \"{model}\" is loaded with {actual} tokens, configured {configured} — overflow risk",
|
|
468
|
+
"repl.context_probe_big": "Context probe: model \"{model}\" supports {actual} tokens — consider \"mma context {actual}\"",
|
|
467
469
|
"repl.show": "show",
|
|
468
470
|
"repl.hide": "hide",
|
|
469
471
|
"repl.resume_hint": "Use /resume <id> or /resume <name> to switch between sessions",
|
package/dist/i18n/ru.json
CHANGED
|
@@ -458,6 +458,8 @@
|
|
|
458
458
|
"repl.agents_label": "Инструкции:",
|
|
459
459
|
"repl.not_found": "не найден",
|
|
460
460
|
"repl.disabled": "отключено (--no-agents-md)",
|
|
461
|
+
"repl.context_probe_small": "Context-проб: модель \"{model}\" загружена с контекстом {actual} токенов, настроено {configured} — риск переполнения",
|
|
462
|
+
"repl.context_probe_big": "Context-проб: модель \"{model}\" поддерживает {actual} токенов — рассмотрите \"mma context {actual}\"",
|
|
461
463
|
"repl.show": "показ",
|
|
462
464
|
"repl.hide": "скрыть",
|
|
463
465
|
"repl.resume_hint": "Используйте /resume <id> или /resume <имя> для переключения между сессиями",
|