matevip
139a478bcd
fix(skill): give CJK-only MCP/ACP names a stable id-based slug
2026-05-14 18:18:15 +08:00
matevip
d1be02985c
fix(agent): collapse model-predicted tool results that the actual tool replaced
2026-05-14 15:44:03 +08:00
matevip
f0f97232a8
fix(agent): pre-tool no-claim system rule + correct write/edit approval docs
2026-05-14 14:59:32 +08:00
matevip
bf3f4af913
fix(agent): structured tool-call replay + final-answer-only content ( #120 )
2026-05-14 07:46:30 +08:00
matevip
2bca55f5b4
fix(tool-guard): friendly duplicate rule-id rejection and client-side name/pattern validation ( #88 )
2026-05-13 09:55:30 +08:00
matevip
aed559d1cf
docs(agent): update head-guard call-site comment to match the order-sensitive forward scan
2026-05-13 09:05:17 +08:00
matevip
1ea90c25cb
fix(agent): make head-orphan repair order-sensitive — a later same-id assistant does not redeem an earlier orphan
2026-05-13 09:03:16 +08:00
matevip
bff57924d6
fix(agent): repair head-side orphan tool responses on the pagination cut
2026-05-13 09:03:10 +08:00
matevip
af56763156
fix(context): keep the latest summary even when its row sits outside the recent window; include summaryId in compact_status
2026-05-13 08:50:42 +08:00
matevip
faf7f98358
feat(context): consume compact_status on the frontend; load history from latest boundary; backfill summaryId
2026-05-13 08:50:35 +08:00
matevip
51b5eceb7d
feat(context): preserve spill markers across compaction phases, persist enriched boundary, broadcast compact_status SSE ( #110 )
2026-05-13 08:50:28 +08:00
matevip
86a6829102
feat(executor): retention sweep and per-conversation purge for tool-result spill files
2026-05-13 08:50:22 +08:00
matevip
7015d1513c
feat(context): anchor the first user message after compaction so long tasks remember the original goal
2026-05-13 08:50:15 +08:00
matevip
37894a6978
fix(context): never compact across a split tool_call ↔ tool_response pair
2026-05-13 08:50:08 +08:00
matevip
ae3d27922d
fix(context): preserve raw older tool results instead of rewriting them to a lossy summary
2026-05-13 08:50:02 +08:00
matevip
8f49b0ec92
fix(executor): spill raw tool result before falling back to inline truncate
2026-05-13 08:49:56 +08:00
matevip
2ef806aa64
chore(deps): bump Spring AI 1.1.5 -> 1.1.6 and Spring AI Alibaba 1.1.2.2 -> 1.1.2.3
2026-05-13 07:39:27 +08:00
matevip
cc5d7d5ad0
fix(agent): auto-include enabled MCP tools in effective allowlist ( #108 )
2026-05-12 21:24:36 +08:00
matevip
aded30bbb1
feat(agent): auto-redirect skill-as-tool calls to readSkillFile
2026-05-12 17:20:19 +08:00
matevip
5789a28e19
feat(skill): boost newly installed skills + teach LLM to widen the catalog search
2026-05-12 17:20:09 +08:00
matevip
0b321dc903
fix(skill): repair zip install losses + persist scripts/refs to canonical store ( #104 )
2026-05-12 17:20:00 +08:00
matevip
691d2b867b
fix(agent): hide disabled agents from the chat picker and reject chat calls against them ( #105 )
2026-05-12 14:53:14 +08:00
matevip
026afa2ba5
feat(wiki): optional JSON Schema on json-format transformations
2026-05-12 14:35:31 +08:00
matevip
465a727be2
feat(wiki): record per-run token usage so operators can see what each template burns
2026-05-12 14:35:17 +08:00
matevip
e911af2192
feat(wiki): JSON output mode — structured transformation output for programmatic downstream
2026-05-12 14:11:02 +08:00
matevip
90936dba4f
feat(wiki): cross-material transformation aggregator — map-reduce all runs of a template into one KB page
2026-05-12 14:10:54 +08:00
matevip
a5b952050c
feat(wiki): reverse-citation extractor — bind synthesis pages to the exact source chunks they cite
2026-05-12 14:10:47 +08:00
matevip
e1759e0e3c
feat(wiki): page-input transformations — run a template against an existing page
2026-05-12 14:10:40 +08:00
matevip
bee39cc420
feat(wiki): page-level embedding so synthesis pages enter semantic search
2026-05-12 14:10:23 +08:00
matevip
8d01efbbb1
fix(agent): sanitize streamed tool-call arguments to valid JSON
2026-05-12 11:25:58 +08:00
matevip
62c9776b01
feat(wiki): clean conversational preamble and code fences from transformation output
2026-05-12 11:25:50 +08:00
matevip
f04ea58c8d
feat(wiki): cancel a running transformation and re-run any past run
2026-05-12 11:25:44 +08:00
matevip
ef6ad1b91a
feat(wiki): expose per-template model picker in the transformations modal
2026-05-12 11:25:38 +08:00
matevip
4098346fb7
fix(wiki): strip file extensions when deriving transformation page slugs
2026-05-12 11:25:32 +08:00
matevip
bae9f68261
feat(wiki): user-defined transformation templates with optional auto-save to synthesis pages
2026-05-12 10:49:58 +08:00
srant1
5f77434953
fix(llm): activate default model after OpenAI OAuth and gate provider checks by enabled
...
OAuth token save now promotes the first available chat model when no usable default exists. Default-model resolution and provider availability checks require Provider.enabled=true alongside credentials, so disabled Providers no longer return stale defaults. /models/enabled drops the single-Provider hard-code so OpenAI OAuth and other enabled chat models surface in selectors. Docker exposes the 1455 PKCE callback via MATECLAW_OAUTH_OPENAI_CALLBACK_BIND_HOST (default 127.0.0.1) and a port mapping; deployment mode stays Host-driven.
2026-05-12 09:58:30 +08:00
matevip
27f614d48a
feat(wiki): enrich chunk embedding input with raw title and structural metadata
2026-05-12 08:44:26 +08:00
matevip
dac3f5be20
fix(extract): trigger OCR when text extraction returns unreadable bytes
2026-05-12 08:44:20 +08:00
matevip
071ccff9bf
feat(tool): add render_html_image to deliver HTML artifacts as native IM images
2026-05-12 07:11:35 +08:00
matevip
b74176ef66
refactor(agent): tighten template pre-binding contract and prune product-assistant skills
2026-05-12 07:11:29 +08:00
matevip
201e510b17
feat(agent): pre-bind skills and tools when applying an agent template
2026-05-12 07:11:19 +08:00
matevip
08fcab12d1
fix(deploy): drop DASHSCOPE_API_KEY requirement so Docker boots with no LLM key configured ( #99 )
2026-05-11 20:44:31 +08:00
matevip
e6fd39f6ba
fix(agent): raise parallel delegation timeout to 300s for thinking models
2026-05-11 17:23:26 +08:00
matevip
f25f910a82
feat(skill): support optional: true frontmatter for off-by-default bundled skills
2026-05-11 17:22:30 +08:00
matevip
6d5fb595d3
fix(agent): harden digital-employee isolation across creation, skill binding and prompt catalog
2026-05-11 16:10:28 +08:00
matevip
9fd5843dda
fix(llm): pin OpenAI-compatible HTTP client to HTTP/1.1 ( #89 )
2026-05-11 11:22:23 +08:00
matevip
3f1f83bbfe
fix(channel): gate WS/long-polling channels on a distributed leader lease ( #85 )
2026-05-11 10:50:00 +08:00
matevip
b7f69dbefe
fix(skill): merge partial PUT into existing row so SKILL.md saves stop 500ing ( #93 )
2026-05-11 07:59:49 +08:00
matevip
9cac7a31d1
refactor(skill-mcp): expose tool display names via ResolvedSkill instead of Map view
2026-05-10 23:46:37 +08:00
czhcc
18148aa724
feat(skill-mcp): expose readable display name alongside prefixed MCP tool names
...
Add a parallel effectiveAllowedToolsDisplay field on the runtime status payload so the SkillMarket detail drawer can render mcp_<server>_<slug>_<hash> with the raw tool name appended in parentheses, while leaving the original prefixed list unchanged for any caller that needs the machine name. McpSkillBridge#decorateToolNameForDisplay reverses a prefixed name via the per-server cached tool list; the frontend prefers the new display field and falls back to effectiveAllowedTools when the field is absent.
2026-05-10 23:30:22 +08:00
