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

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>

docs/architecture/COMPANION_BRAIN_ARCHITECTURE.md

@@ -1993,6 +1993,18 @@ Chosen over the deep-research pipeline (`/api/v1/research/start-sync`) because:
 
 Failures throw explicitly (0 feeds or 0 articles) — the runner catches and injects a "research failed" `ResolvedInput` with the error message so the planner doesn't hallucinate URLs.
 
+#### Server-Side Research (mana-ai, ab v0.6)
+
+The `mana-ai` background runner mirrors the client-side research pre-step. `NewsResearchClient` (`services/mana-ai/src/planner/news-research-client.ts`) calls `mana-api`'s `POST /api/v1/news-research/discover` + `/search` directly over the Docker network (`MANA_API_URL`). The same `RESEARCH_TRIGGER` regex is used; when it matches, results are injected as `ResolvedInput { id: '__web-research__' }` before the planner prompt is built.
+
+Key differences from the client-side path:
+- No SvelteKit context — pure HTTP fetch.
+- 15s timeout on discover, 30s on search (tighter than client because the tick has a 60s cadence).
+- Graceful null on any failure — the planner just runs without research context; tick doesn't crash.
+- No `discoverByQuery` / `searchFeeds` module imports — the client ships those from `@mana/shared-rss`; the server calls the API endpoints which wrap the same logic.
+
+This makes the Recherche-Agent and Today-Agent fully autonomous (no browser tab required). `research_news` is also registered as a proposable tool so the planner can explicitly request additional research as a PlanStep.
+
 ### 23.3 Kontext Auto-Inject
 
 The user's `kontextDoc` singleton is automatically appended to every planner call as a standing-context `ResolvedInput`, unless the mission already links it as an explicit input. Decrypted client-side only — the server-side mana-ai runner skips this (encryption barrier; needs a Key-Grant for server access).

docs/plans/README.md

@@ -33,12 +33,13 @@ each one laying foundations the next one relies on:
 | Plan | Status | Scope |
 |---|---|---|
 | [`ai-mission-key-grant.md`](./ai-mission-key-grant.md) | ✅ Shipped | Per-mission RSA-wrapped key grant so `mana-ai` can decrypt allowlisted encrypted records when user opts in. |
-| [`multi-agent-workbench.md`](./multi-agent-workbench.md) | ✅ Shipped | Identity-aware Actor + named AI agents owning missions + per-agent policy + scene lens. |
+| [`multi-agent-workbench.md`](./multi-agent-workbench.md) | ✅ Shipped | Identity-aware Actor + named AI agents owning missions + per-agent policy + scene lens. 28 tools across 11 modules including server-side web-research. |
+| [`workbench-templates.md`](./workbench-templates.md) | ✅ T1 Shipped | Generalised templates: 3 agent-templates + 3 non-AI workbench starter-kits. Seed-handler registry for per-module data seeding. |
 | [`team-workbench.md`](./team-workbench.md) | 📝 Forward-looking | TeamSpace with membership, team-encrypted records, admin lens on team members. Reuses Actor.principalId + key-wrapping patterns from the two above. |
 
 Cross-references:
 
-- Architecture narrative: [`docs/architecture/COMPANION_BRAIN_ARCHITECTURE.md`](../architecture/COMPANION_BRAIN_ARCHITECTURE.md) §20 (AI Workbench base), §21 (Mission Grants), §22 (Multi-Agent)
+- Architecture narrative: [`docs/architecture/COMPANION_BRAIN_ARCHITECTURE.md`](../architecture/COMPANION_BRAIN_ARCHITECTURE.md) §20 (AI Workbench base), §21 (Mission Grants), §22 (Multi-Agent), §23 (Reasoning Loop + Research + Debug)
 - Non-plan ideas backlog: [`docs/future/AI_AGENTS_IDEAS.md`](../future/AI_AGENTS_IDEAS.md)
 - Service-internal notes: [`services/mana-ai/CLAUDE.md`](../../services/mana-ai/CLAUDE.md)
 - Webapp-internal notes: [`apps/mana/CLAUDE.md`](../../apps/mana/CLAUDE.md) → "AI Workbench" section