till/managarten
1
0
Fork 0
mirror of https://github.com/Memo-2023/mana-monorepo.git synced 2026-05-17 05:59:39 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 4

docs: update tool coverage table + server-side research + templates

Browse source 
Catches up all docs with the current state of the AI tool system.

services/mana-ai/CLAUDE.md:
- New v0.6 status section documenting NewsResearchClient,
  pre-planning research injection, config.manaApiUrl, and the full
  28-tool / 11-module inventory (17 propose + 11 auto).

apps/mana/CLAUDE.md:
- New "Tool Coverage" table in the AI Workbench section listing all
  tools per module with their policy (propose vs auto).
- New "Templates" subsection documenting the two-section gallery
  (agent vs workbench templates), the seed-handler registry, and
  the current handlers (meditate, habits, goals).
- Architecture cross-reference updated to include §23.

docs/architecture/COMPANION_BRAIN_ARCHITECTURE.md:
- §23.2 gains a "Server-Side Research (mana-ai, ab v0.6)" subsection
  explaining how NewsResearchClient mirrors the client-side research
  pre-step: same endpoints, same trigger regex, but HTTP-direct from
  the Docker network instead of SvelteKit-internal.

docs/plans/README.md:
- workbench-templates.md added to the roadmap table (T1 shipped).
- Multi-agent description updated to mention 28 tools + server-side
  web-research.
- Architecture cross-reference includes §23.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
Till JS 2026-04-16 12:35:40 +02:00
parent 2ead0f36df
commit d83fc370a0
4 changed files with 56 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

31
apps/mana/CLAUDE.md
View file

@ -183,7 +183,36 @@ The companion is a **second actor** that works alongside the human in every modu
- **Scene lens** — workbench scenes can bind to an agent via `scene.viewingAsAgentId` (context menu → "An Agent binden…"). Pure UI lens, not a data-scope change. `SceneAppBar` shows the agent avatar on bound scene tabs.
- **Workbench timeline** — `/ai-workbench` renders every AI-attributed event grouped by mission iteration with per-**agent** filter, per-module, per-mission. Each bucket header shows agent avatar + name + mission title. Per-bucket **Revert button** undoes the iteration's writes via `data/ai/revert/` (TaskCreated → delete, TaskCompleted → uncomplete, etc., newest-first). Separate **"Datenzugriff"** tab exposes the server-side decrypt audit (for missions with Key-Grants).
Full architecture (Planner prompt + parser in `@mana/shared-ai`, server-side runner, Postgres actor column, materialized snapshots, Multi-Agent gating, Prometheus metrics + status.mana.how integration): [`docs/architecture/COMPANION_BRAIN_ARCHITECTURE.md`](../../docs/architecture/COMPANION_BRAIN_ARCHITECTURE.md) §20 (AI Workbench) + §21 (Mission Grants) + §22 (Multi-Agent Workbench).
### Tool Coverage (28 tools, 11 modules)
Agents interact with the app through tools — each one either auto (executes silently during reasoning) or propose (creates a Proposal card the user must approve). Canonical list in `@mana/shared-ai/src/policy/proposable-tools.ts`; server-side definitions in `services/mana-ai/src/planner/tools.ts`; webapp auto-tool list in `src/lib/data/ai/policy.ts`.
| Module | Propose | Auto |
|--------|---------|------|
| todo | `create_task`, `complete_task`, `complete_tasks_by_title` | `get_task_stats`, `list_tasks` |
| calendar | `create_event` | `get_todays_events` |
| notes | `create_note`, `update_note`, `append_to_note`, `add_tag_to_note` | `list_notes` |
| places | `create_place`, `visit_place` | `get_places`, `location_log` |
| drink | `undo_drink` | `get_drink_progress`, `log_drink` |
| food | — | `nutrition_summary`, `log_meal` |
| news | `save_news_article` | — |
| news-research | `research_news` | — |
| journal | `create_journal_entry` | — |
| habits | `create_habit`, `log_habit` | `get_habits` |
| contacts | `create_contact` | `get_contacts` |
**Server-side web-research**: mana-ai calls mana-api's `/api/v1/news-research/discover` + `/search` directly before the planner prompt is built (pre-planning injection). Missions with research-keyword objectives get real article URLs + excerpts injected as a synthetic ResolvedInput. See `services/mana-ai/src/planner/news-research-client.ts`.
### Templates
Pre-configured starter-kits at `/agents/templates` — two sections:
- **Agent-Templates** (with AI): Recherche-Agent, Kontext-Agent, Today-Agent
- **Workbench-Templates** (no AI): Calmness, Fitness, Deep Work
Each template bundles: optional agent + optional scene layout + optional starter missions (paused) + optional per-module seeds. Template shape: `WorkbenchTemplate` in `@mana/shared-ai/src/agents/templates/types.ts`. Applicator: `src/lib/data/ai/agents/apply-template.ts`. Seed-handler registry: `src/lib/data/ai/agents/seed-registry.ts` — modules register via side-effect imports in `missions/setup.ts`. Current handlers: meditate, habits, goals. Plan: [`docs/plans/workbench-templates.md`](../../docs/plans/workbench-templates.md).
Full architecture (Planner prompt + parser in `@mana/shared-ai`, server-side runner, Postgres actor column, materialized snapshots, Multi-Agent gating, server-side web-research, Prometheus metrics + status.mana.how integration): [`docs/architecture/COMPANION_BRAIN_ARCHITECTURE.md`](../../docs/architecture/COMPANION_BRAIN_ARCHITECTURE.md) §20 (AI Workbench) + §21 (Mission Grants) + §22 (Multi-Agent Workbench).
## Reference Documents