Commit Graph

12 Commits

Author SHA1 Message Date
matevip
5e6764ecf1 fix(skill): tool-gate whitelist + markdown link host normalization
When an agent had any skill bound, the runtime tool gate was silently
hiding @Tool beans that aren't declared in any skill manifest, even
though the global system prompts (SOUL.md / "Web Search Capability" /
"File Reading Guidelines") explicitly tell the LLM these tools are
available. Result: the model would call search / renderDocx / read_file
/ etc., hit "Tool not found", then either give up or fall back to
unhelpful behaviour (e.g. dumping markdown text instead of producing a
.docx download).

This commit:

- Adds universally-promised, agent-wide tools to SYSTEM_LEVEL_TOOLS so
  they bypass the manifest restriction: document/media generation
  (renderDocx*, image_generate, music_generate, video_generate),
  global capability tools the system prompt mentions (search,
  browser_use, read_file / write_file / edit_file /
  execute_shell_command, detect_file_type, extract_*_text,
  readMateClawDoc), skill discovery siblings (listSkillFiles,
  listAvailableSkills), and the delegate triplet (delegateToAgent,
  delegateParallel, listAvailableAgents).

- Fixes 5 entries in the prior whitelist whose names did not match
  any real @Tool bean and were therefore silently dead:
    read_workspace_file   -> read_workspace_memory_file
    write_workspace_file  -> write_workspace_memory_file
    list_workspace_files  -> list_workspace_memory_files
    delegate_agent        -> delegateToAgent
    datetime              -> getCurrentDate / getCurrentDateTime / getCurrentTime
  Also adds the missing edit_workspace_memory_file.

- In the chat markdown renderer, strips any hallucinated
  https?://<host> prefix from /api/v1/files/generated/<id> download
  links before building the <a href>. Multiple LLMs have been
  observed prepending bogus hosts when echoing tool-returned download
  URLs back to the user, breaking the click. One-line defensive
  normalization independent of which model is in use.

Verified end-to-end on a previously-broken agent: search / browser_use
/ execute_shell_command / renderDocx all dispatch correctly now and
the final markdown link is a clean same-origin path. 36 whitelist
entries cross-checked against real @Tool method names.
AgentBindingServiceTest green.
2026-05-01 12:24:25 +08:00
matevip
8ce7e7d26d feat(ui): agent integration of SkillIcon + SkillIconPicker 2026-05-01 10:58:15 +08:00
matevip
8ad0fca4ef fix(memory): load structured/*.md by encoding path segments 2026-04-29 15:03:01 +08:00
matevip
96b89414f7 feat(memory): enable Memory Browser tab + audit fixes 2026-04-21 15:28:31 +08:00
matevip
6ae221b38d feat(memory): batch 3 — Grafana dashboard + 3 Low fixes + issues cleanup 2026-04-21 15:28:21 +08:00
matevip
e42444f50a feat(ui): batch 2 — Facts Tab + Contradiction Inbox + Trust Bar 2026-04-21 15:28:16 +08:00
matevip
c252cf50a4 docs: update README title and tagline 2026-04-21 09:27:42 +08:00
matevip
d9d677e762 refactor(ui): complete Memory view redesign — kill EP, match design system 2026-04-21 04:59:01 +08:00
matevip
bf4cebb7b3 feat(memory): dream-v2 D3 — Diff viewer + SSE + Focused Dream dialog 2026-04-21 04:58:43 +08:00
matevip
30b9912ded refactor(ui): redesign Memory view — follow mc-page-shell design system 2026-04-21 04:58:39 +08:00
matevip
8ecc6d10ca feat(memory): dream-v2 D2 — Morning Card + HiL (Confirm/Edit) 2026-04-21 04:58:33 +08:00
matevip
90067d1c9f feat(memory): dream-v2 D1 — Memory Timeline view (frontend + backend) 2026-04-21 04:58:26 +08:00