prism-mcp-server 20.21.14 → 20.21.16

package/README.md

@@ -158,6 +158,42 @@ or by re-enabling after each run.
 <details>
 <summary>Release history (optional)</summary>
 
+## What's New in v20.21.16
+
+### Conversations: checked on your device, and free with an account
+
+- A local answer to a conversation is checked on your device before it is
+  served. If the check fails or cannot run, the answer goes to the cloud when
+  your plan and settings allow it, and is withheld otherwise.
+- On Pro and Team, a local answer that passes is also confirmed by Synalux, on
+  a copy pseudonymized on your device; the mapping stays on your device. With
+  an image attached, or with the cloud off, the checked local answer is served.
+- When the on-device screen's 4b is unsure about a conversation, the 9b reads
+  it again before the conversation is refused or sent to the cloud.
+- Everything local is free with no account and no cap on the local model size.
+  A free Synalux account adds multi-turn `prism_infer` and 20 cloud answers a
+  day. Pro ($15/month) and Team ($20/seat/month) are Prism's own plans.
+- Optional You.com web search: set `YDC_API_KEY` to enable
+  `youcom_web_search` (contributed by @mouse-value-add in #227).
+- Fixed: the 2b and 4b no longer answer Prism's own requests as a tool router.
+  A system prompt built into those models had turned some answers into tool
+  calls and made the on-device screen refuse some ordinary requests.
+
+## What's New in v20.21.15
+
+### Follow-ups: see what runs locally, and a stricter screen when the classifier is down
+
+- `local_savings` (and `prism savings`) shows a line for follow-ups, calls that
+  carried your conversation. It covers how many the local model answered (and
+  how many of those the 9b answered), how many the on-device screen refused and
+  at which stage, how many your plan's multi-turn limits refused, and how many
+  went to the cloud.
+- A short chat answer ("16", "Yes") counts as an answer. Before, the quality
+  gate treated it as empty, and a paid plan re-asked the cloud.
+- When the on-device classifier fails every read on a follow-up, the follow-up
+  is no longer answered on a keyword check alone. It goes to the cloud if your
+  plan allows it, and is refused otherwise.
+
 ## What's New in v20.21.14
 
 ### Skills load when they help, and you can see why they loaded
@@ -1135,7 +1171,7 @@ The free tier runs entirely on your machine. Paid tiers add cloud sync through t
 | Inference | Local Ollama models | Local models + Gemini 3.6 Flash fallback |
 | API keys required | None | Synalux subscription key |
 | Web search / scrape | Not included | Via Synalux portal (provider keys server-side) |
-| What leaves your machine | Nothing | Memory text, file paths, search queries, and inference prompts/drafts when their cloud feature is used, sent to the portal over TLS. Cloud memory writes are PHI-redacted; inference and route requests are transient. |
+| What leaves your machine | Nothing | Memory text, file paths, search queries, and inference prompts/drafts when their cloud feature is used, sent to the portal over TLS. Cloud memory writes have personal identifiers redacted; inference and route requests are transient. |
 | Works offline | ✅ | Local features yes; sync/cloud no |
 
 **Handling sensitive data.** Cloud memory writes pass through automatic
@@ -1144,8 +1180,7 @@ and clinical identifiers are stripped before storage). Cloud inference and
 route correction send the request over TLS for processing and do not store it
 as Prism memory. The **local tier** (no Synalux key) is the air-gap;
 `route_guard: "local"` only skips the route correction, and `cloud_fallback`
-and `verify` govern the other two channels. **Enterprise** includes a HIPAA
-Business Associate Agreement.
+and `verify` govern the other two channels.
 
 ---
 
@@ -1171,8 +1206,8 @@ and telemetry still contact the portal and carry neither.
 |---|---|---|---|---|---|---|
 | Qwen3.5-4B Q4_K_S | `prism-coder:2b` | 3.3 GB | ✅ | 100% | On-device / lowest RAM (4.5 GiB free) | Free |
 | Qwen3.5-4B Q4_K_M | `prism-coder:4b` | 3.5 GB | ✅ | 100% | Verifier (5.2 GiB free) | Free |
-| Qwen3.5-9B (LoRA) | `prism-coder:9b` | 6.7 GB | ✅ | 95.7%² | Default router / workhorse (9 GiB free) | Standard+ |
-| Qwen3.5-27B (LoRA) | `prism-coder:27b` | 16.8 GB | — | 100% | Complex code / quality (21 GiB free) | Advanced+ |
+| Qwen3.5-9B (LoRA) | `prism-coder:9b` | 6.7 GB | ✅ | 95.7%² | Default router / workhorse (9 GiB free) | Free |
+| Qwen3.5-27B (LoRA) | `prism-coder:27b` | 16.8 GB | — | 100% | Complex code / quality (21 GiB free) | Free |
 
 ¹ Self-run on a narrow 115-case MCP tool-selection suite, `temperature: 0`,
 measured through the call path `prism_infer` actually uses (`/api/chat`, each
@@ -1284,74 +1319,64 @@ an image is never escalated; screenshots stay on this device.
 
 ## Why Prism Coder
 
-### vs AI coding assistants
-
-Product capabilities and plans change frequently. The comparison below is
-intentionally limited to publicly documented differences; it is not a claim
-that another product lacks an unlisted feature.
-
-Legend: ✅ documented, ◐ conditional or plan-dependent, — not compared, ? verify
-with the provider.
-
-| Capability | Prism Coder | GitHub Copilot | Cursor | Amazon Q Developer |
-|---|:---:|:---:|:---:|:---:|
-| Local/open-weight inference | ✅ | ◐ | ◐ | ◐ |
-| Offline workflow | ✅ | ◐ | ? | ? |
-| Cross-session memory | ✅ | ◐ ([docs](https://docs.github.com/en/copilot/reference/copilot-cli-reference/cli-command-reference)) | ◐ | ◐ |
-| MCP integration | ✅ | ✅ ([docs](https://docs.github.com/en/copilot/reference/copilot-cli-reference/cli-command-reference)) | ✅ ([pricing](https://cursor.com/en-US/pricing)) | ◐ |
-| Local-first model routing | ✅ | ◐ | ◐ | ◐ |
-| Session drift and grounding checks | ✅ | — | — | — |
-| Setup surface | ✅ five hosts | ✅ CLI/IDE | ✅ editor/agents | ✅ IDE/CLI ([overview](https://aws.amazon.com/q/developer/build/)) |
-| Pricing model | ✅ Synalux tiers | ◐ | ◐ ([pricing](https://cursor.com/en-US/pricing)) | ✅ free + $19 Pro ([pricing](https://aws.amazon.com/q/developer/pricing/)) |
+**Your local data stays on your device.** Memory, local models and the answer
+check run on your machine, with no account. Data leaves it only when you turn
+on a cloud feature, and then only what that feature needs; see
+[Local-first and privacy](#local-first-and-privacy).
 
-Prism-specific compliance, contractual, and pricing terms are documented in
-the Synalux service agreement. Do not infer a competitor's HIPAA, BAA, or data
-handling status from this table.
+- **Free without an account**, with no cap on the local model size.
+- **Memory and models in one MCP server:** persistent memory, a knowledge graph
+  and automatic routing to local Ollama models, with cloud answers only when
+  your plan and settings allow them.
+- **Checked answers:** a multi-turn local answer is checked on your device
+  before it is served.
 
-### vs local AI / memory tools
+### How it compares
 
-| Feature | Prism Coder | Ollama | LM Studio | Mem0 | Zep |
-|---|:---:|:---:|:---:|:---:|:---:|
-| Local inference cascade | ✅ | ✅ runtime | ✅ app | — | — |
-| Cloud fallback | ✅ optional | — | ◐ provider-dependent | ◐ | ◐ |
-| Persistent memory | ✅ | — | ◐ project context | ✅ | ✅ |
-| Knowledge/tool integration | ✅ MCP + ingestion | ◐ APIs | ◐ integrations | ✅ SDK/API | ✅ SDK/API |
-| MCP server | ✅ native | — | ◐ client integration | ◐ client integration | ◐ client integration |
+Prices from each vendor's pricing page, checked on 2026-09-26. Plans change
+often; follow the links for current terms. The table compares published plans
+only; it is not a claim that a product lacks an unlisted feature.
 
-### Pricing
-
-Prism's current published tiers are listed below. Competitor pricing is
-usage- and plan-dependent, so consult the provider directly: [GitHub
-Copilot](https://github.com/features/copilot/plans), [Cursor](https://cursor.com/en-US/pricing),
-and [Amazon Q Developer](https://aws.amazon.com/q/developer/pricing/).
+| Product | What it is | Free | Individual | Team, per seat |
+|---|---|---|---|---|
+| **Prism Coder** | Local memory and model routing (MCP), optional cloud | Everything local, no account | Pro $15/mo | $20, 2-seat minimum |
+| [GitHub Copilot](https://docs.github.com/en/copilot/get-started/plans) | Coding assistant | Limited | Pro $10, Pro+ $39, Max $100 | Business $19, Enterprise $39 |
+| [Cursor](https://cursor.com/pricing) | AI code editor | Hobby, limited | Pro $20 | Teams $40 |
+| [Windsurf](https://devin.ai/blog/windsurf-pricing-plans) | AI code editor | Free plan | Pro $20, Max $200 | Teams $40 |
+| [Claude](https://claude.com/pricing) | Assistant; Claude Code from Pro | Free (no Claude Code) | Pro $20, Max from $100 | Team $25 ($20 billed annually) |
+| [Amazon Q Developer](https://aws.amazon.com/q/developer/pricing/) | Coding assistant | 50 agentic requests a month | Pro $19 per user | -- |
+| [Ollama](https://ollama.com/pricing) | Local model runtime, cloud models | Local models, starter cloud credits | Pro $20 ($60 of cloud credits), Max $100 | Team $500/mo, unlimited users |
+| [Mem0](https://mem0.ai/pricing) | Memory API | 10,000 adds, 1,000 retrievals a month | Starter $19; graph memory from Pro $249 | -- |
+| [Supermemory](https://supermemory.ai/pricing/) | Memory API | $5 of credits a month | Pro $19 (3 seats), Max $100 | -- |
+| [Zep](https://www.getzep.com/pricing/) | Memory (knowledge graph) | 10,000 credits a month | Flex $125 | -- |
+| [Pieces](https://pieces.app/pricing) | On-device developer memory | None (7-day trial) | Pro $18.99 | Enterprise $22.99 |
+| [MemPalace](https://www.mempalace.net/) | Open-source local memory (MIT) | Free | -- | -- |
 
 ---
 
 ## Plans
 
-All on-device models are free to run locally via Ollama on every tier. A subscription gates **cloud** features, higher automatic-routing ceilings, and increased limits. On-device models run through your Ollama regardless of plan; the ceiling applies only to cloud inference and automatic `prism_infer` routing.
+Everything that runs on your machine is free, with no account and no cap on the local model. A free Synalux account adds what needs Synalux; Pro and Team add cloud features and higher limits.
 
-| | **Free** | **Standard** $19/mo | **Advanced** $49/mo | **Enterprise** $99/mo |
+| | **Free** (no account) | **Free account** | **Pro** $15/mo | **Team** $20/seat/mo |
 |---|---|---|---|---|
-| Seats | 1 | 1 | up to 5 | up to 25 |
-| Automatic `prism_infer` ceiling | up to 4b | up to 9b | up to 27b | up to 27b |
-| Cloud inference | -- | ✅ | ✅ | ✅ (priority) |
-| Cloud Coder (Web IDE) | -- | ✅ | ✅ | ✅ (priority) |
-| Cloud search | -- | ✅ | ✅ | ✅ |
-| Max output tokens | 512 | 1,024 | 2,048 | 4,096 |
-| Cloud fallback | -- | Gemini 3.6 Flash | Gemini 3.6 Flash | Gemini 3.6 Flash (priority) |
-| Multi-turn `prism_infer` (conversation carried across calls) | -- | 12 turns / 32k chars | 20 turns / 64k chars | 30 turns / 96k chars |
-| Grounding verifier (fact-check AI output) | -- | ✅ | ✅ | ✅ |
-| Memory sync (cloud) | -- | ✅ | ✅ | ✅ |
-| Knowledge / session memory | limited | unlimited | unlimited | unlimited |
-| Analytics dashboard | -- | ✅ | ✅ | ✅ |
-| HIPAA BAA | -- | -- | -- | ✅ |
-
-Paid plans start with a 14-day trial. No card is required to start; add payment
+| Local models via Ollama (automatic `prism_infer` routing) | up to 27b | up to 27b | up to 27b | up to 27b |
+| Local memory, knowledge graph, dashboard, MCP tools | ✅ | ✅ | ✅ | ✅ |
+| Multi-turn `prism_infer`, with the on-device answer check | -- | 12 turns / 32k chars | 12 turns / 32k chars | 20 turns / 64k chars |
+| Cloud answers (Gemini 3.6 Flash; a free account's conversations: 3.5 Flash-Lite) | -- | 20/day | 1,000/month (500 in the trial; 200/day max) | 1,500/month per seat (200/day max) |
+| Cloud confirmation of a local answer (on a pseudonymized copy) | -- | -- | ✅ | ✅ |
+| Grounding verifier (fact-check AI output; counts as a cloud answer) | -- | -- | ✅ | ✅ |
+| Cloud search | -- | -- | 1,000/month (150/day max) | 1,500/month per seat (150/day max) |
+| Memory sync (cloud) | -- | -- | ✅ | ✅ |
+| Full skill library, super-skills, agent routing, account and team skills | -- | -- | ✅ | ✅ |
+| Seats | 1 | 1 | 1 | 2 or more |
+
+Pro starts with a 14-day trial. No card is required to start; add payment
 details before the trial ends to continue, or the subscription cancels
-automatically and local Prism Free remains available. View the exact deadline
-under **⚙ Account & Settings → Account** in `prism dashboard`. 25+ seats:
-[contact sales](https://synalux.ai/support)
+automatically and Prism Free remains available. View the exact deadline under
+**⚙ Account & Settings → Account** in `prism dashboard`. A cloud confirmation
+counts as one of the day's cloud answers. To start a Team plan:
+[contact sales](https://synalux.ai/support).
 
 ---
 
@@ -1445,11 +1470,11 @@ It is paid because it cannot run without Synalux behind it:
 ```typescript
 // Call 1
 prism_infer({ prompt: "My project codename is Nightjar. Reply OK.", mode: "chat" })
-// → "OK"                                  (local 9b, $0)
+// → "OK"                                  (local model, $0)
 
 // Call 2 — the model never saw call 1
 prism_infer({ prompt: "What is my codename? One word.", mode: "chat" })
-// → "I don't have that information."      (local 9b, correct and useless)
+// → "I don't have that information."      (local model, correct and useless)
 
 // Call 3 — a coding follow-up with no thread
 prism_infer({ prompt: "Now add a timeout parameter to it.", mode: "code" })
@@ -1474,7 +1499,7 @@ prism_infer({
     prompt: "What is my codename? One word.",
     mode: "chat",
 })
-// → "Nightjar"                            (local 9b, $0; history_turns: 2)
+// → "Nightjar"                            (local model, $0; history_turns: 2)
 
 prism_infer({
     messages: [
@@ -1484,7 +1509,7 @@ prism_infer({
     prompt: "Write the one-line call that stores its result in n.",
     mode: "code",
 })
-// → "n = countActiveUsers(data)"          (local 9b, $0)
+// → "n = countActiveUsers(data)"          (local model, $0)
 
 // A turn the on-device screen finds uncertain, alone or in context, is not
 // served locally: it goes to Synalux cloud on a paid plan, or is refused with
@@ -1501,9 +1526,14 @@ prism_infer({
 // → "SYN-4471"                            (Gemini 3.6 Flash; used_cloud: true)
 ```
 
-Measured on the real 9b through the real handler with cloud off: 7 of 12
-benign follow-ups are served locally, the rest refuse and name the reason;
-every injection variant the reviewers built refuses.
+Measured on 20.21.14 through the real handler, on the local 9b with cloud off:
+- Every benign follow-up that reached the local 9b with its conversation
+  attached was answered correctly. Without the conversation, most answers were
+  invented.
+- The on-device screen still refuses too many benign follow-ups: 11 of 26 in
+  our tests. Each refusal names its reason. Cutting these false refusals is
+  the current work.
+- Every reserved multi-turn probe was refused before any generation (11 of 11).
 </details>
 
 | Mode | Think | Model | Use case |
@@ -1556,6 +1586,10 @@ host, or run `prism savings` from a terminal — `--period all|month|week|sessio
     prism-coder:9b: 41 call(s), ~505K tokens
     prism-coder:4b: 12 call(s), ~4.8K tokens
 
+  Follow-ups with your conversation:
+    12 answered locally (12 by the 9b) · 2 refused by the on-device screen · 0 sent to cloud
+    Refusals by stage: follow-up alone 1 · turns together 1
+
   Counts tokens a local model handled instead of your cloud model. On the token
   axis, the token count is measured — a floor, with known undercounts listed
   when present. On the displacement axis, prism cannot observe the call your
@@ -1577,6 +1611,21 @@ the same tokens; and most users are on flat plans where a currency figure means
 nothing at all. Tokens are the one unit prism measured itself. If you know your
 own effective rate, multiply — the split is printed for exactly that reason.
 
+**Follow-ups.** Calls that carried your conversation (`messages`) get their
+own line:
+- how many the local model answered, and how many of those the 9b answered;
+- how many the on-device screen refused, and at which stage (the follow-up read
+  alone, an earlier turn read alone, or the turns read together);
+- how many went to the cloud.
+
+A refused follow-up went back to your host instead of being answered locally,
+so this line shows how much of your follow-up work local serving actually
+took. Refusals because your plan does not include multi-turn history, or the
+history is over its cap, are counted on their own line. They are recorded only
+when the host asks for a report (`escalation: "report"`); otherwise they fail
+before anything is recorded. The line appears for the `week`, `month`, `all`
+and `--days` views, which read the durable ledger.
+
 Refused calls are excluded, the VS Code panel-playground share is disclosed
 separately, and the known sources of undercount are listed inline rather than
 left implicit — so the durable (`week`/`month`/`all`/`--days`) headline is a
@@ -1682,10 +1731,10 @@ command surface, safety model, and verified acceptance cases.
   <img src="docs/scm_search_cli.jpg" alt="prism search — semantic code search with relevance scores" width="500" />
 </p>
 
-### `prism review` — AI code review with HIPAA checks
+### `prism review` — AI code review with security and privacy checks
 
 <p align="center">
-  <img src="docs/scm_review_cli.jpg" alt="prism review — AI code review with security and HIPAA findings" width="400" />
+  <img src="docs/scm_review_cli.jpg" alt="prism review — AI code review with security and privacy findings" width="400" />
 </p>
 
 ### `prism scan` — security scanner for secrets, Dockerfiles, licenses
@@ -1745,9 +1794,9 @@ AI chat, voice input, SOAP note generator, team collaboration, and video calls
 <summary>Feature details</summary>
 
 - **AI**: Chat participant (`@synalux`), multi-agent pipeline, voice input, model switching, 10 tones
-- **Clinical**: SOAP note generator, role-based access, document signing, patient board
+- **Clinical**: SOAP note generator, role-based access, document signing
 - **Collaboration**: Team chat, DMs, video calls, customer board, visual builder, DevContainers
-- **Privacy**: Local Ollama by default. `preferLocal=true` tries local first. Enterprise BAA available.
+- **Privacy**: Local Ollama by default. `preferLocal=true` tries local first.
 </details>
 
 ### Prism AAC
@@ -1846,25 +1895,37 @@ It reads `~/.prism-mcp/data.db` and POSTs entries to the portal. Ledger entries
 
 | Feature | Details |
 |---------|---------|
-| Local inference | Direct Ollama use is unrestricted; automatic `prism_infer` routing selects up to 4B |
+| Local inference | Direct Ollama use is unrestricted; automatic `prism_infer` routing selects up to 27B, as far as your machine's memory allows |
 | Session memory | Persistent sessions, handoffs, ledger — all local SQLite |
 | Knowledge search | Semantic search across session history |
 | Skills | All skills available locally (run `sync-skills.sh` to populate) |
 | Drift detection | Server-side GATE 5 reminders |
 
-### Paid (Synalux subscription)
+### Free account (Synalux sign-in, $0)
 
 Everything in Free, plus:
 
 | Feature | Details |
 |---------|---------|
-| Model ceiling | Automatic `prism_infer` routing up to 27B + Gemini 3.6 Flash fallback when local is unavailable |
+| Multi-turn `prism_infer` | Conversation history carried across calls; each local answer is checked on your device before it is served |
+| Cloud answers | 20 a day from Gemini when a local answer cannot be used: 3.5 Flash-Lite in a conversation, 3.6 Flash for a single prompt |
+
+Sign in from `prism dashboard` under **Account**.
+
+### Pro ($15/mo) and Team ($20/seat/mo, 2-seat minimum)
+
+Everything in the free account, plus:
+
+| Feature | Details |
+|---------|---------|
+| Cloud answers | 1,000 a month (Pro; 500 during the 14-day trial) or 1,500 a month per seat (Team), at most 200 a day, from Gemini 3.6 Flash; a cloud confirmation counts as one |
+| Answer confirmation | A local answer that passed the on-device check is also confirmed by Synalux, on a copy with names, dates and other identifiers replaced on your device |
 | Skill routing | Portal resolves which skills to load based on your project and prompt |
 | Cross-device memory | Supabase cloud sync — sessions survive across machines |
 | Grounding verifier | L3 NLI verification on model outputs |
-| Team features | Multi-agent Hivemind, workspace collaboration |
+| Team features | Multi-agent Hivemind, workspace collaboration, account and team skills |
 
-The paid tier adds **intelligent routing** — the Synalux portal determines which skills are relevant to your current project and prompt, so your agent gets domain expertise (stripe patterns, training protocols, clinical standards) instead of loading everything. Free users with the repo can run `sync-skills.sh` to populate all skills locally; paid routing adds project-aware and prompt-aware selection.
+The paid tiers add **intelligent routing** — the Synalux portal determines which skills are relevant to your current project and prompt, so your agent gets domain expertise (stripe patterns, training protocols, clinical standards) instead of loading everything. Free users with the repo can run `sync-skills.sh` to populate all skills locally; paid routing adds project-aware and prompt-aware selection.
 
 - Contributions require signing the [CLA](./CLA.md).
 - "Prism" and "Synalux" are trade names of Synalux LLC; the Apache license does

package/dist/config.js

@@ -14,6 +14,9 @@ import { fileURLToPath } from "node:url";
  *                            is needed. Get one at https://brave.com/search/api/
  *   GOOGLE_API_KEY         — (optional) API key for Google AI Studio / Gemini. Enables paper analysis.
  *   BRAVE_ANSWERS_API_KEY  — (optional) API key for Brave Answers (AI grounding). Enables brave_answers tool.
+ *   YDC_API_KEY            — (optional) API key for You.com Search. When set, the optional
+ *                            youcom_web_search tool is registered as an alternative web search
+ *                            provider. Get one at https://you.com/platform/api-keys
  *   SUPABASE_URL           — (optional) Your Supabase project URL. Enables session memory tools.
  *   SUPABASE_KEY           — (optional) Your Supabase anon/service key. Enables session memory tools.
  *   PRISM_USER_ID          — (optional) Unique tenant ID for multi-user Supabase instances.
@@ -72,6 +75,18 @@ export const SEMANTIC_SCHOLAR_API_KEY = process.env.SEMANTIC_SCHOLAR_API_KEY;
 if (!BRAVE_ANSWERS_API_KEY && process.env.PRISM_DEBUG_LOGGING === "true") {
     console.error("Warning: BRAVE_ANSWERS_API_KEY environment variable is missing. Brave Answers tool will be unavailable.");
 }
+// ─── Optional: You.com Search API Key ─────────────────────────
+// Used by the youcom_web_search tool. When set, registers an optional
+// web search tool that uses You.com's search API.
+// Get a key at https://you.com/platform/api-keys
+// You.com's own docs use YDC_API_KEY: https://you.com/docs/using-the-api/authentication
+// Trimmed, and blank means unset: a whitespace-only value would otherwise
+// register the tool while every call reports the key as not configured.
+export const YDC_API_KEY = process.env.YDC_API_KEY?.trim() || undefined;
+if (process.env.PRISM_DEBUG_LOGGING === "true") {
+    // Log presence or absence — non-critical, just informational
+    console.error(`[Prism] You.com search: ${YDC_API_KEY ? "configured" : "not configured (youcom_web_search tool disabled)"}`);
+}
 // ─── Optional: Voyage AI Embeddings ──────────────────────────
 // Set embedding_provider=voyage to enable. Requires VOYAGE_API_KEY.
 export const VOYAGE_API_KEY = process.env.VOYAGE_API_KEY; // embedding_provider=voyage

package/dist/dashboard/accountRouter.js

@@ -1,4 +1,5 @@
 import { invalidateEntitlements } from "../utils/entitlements.js";
+import { clearInferencePolicies } from "../utils/inferencePolicy.js";
 import { getSynaluxJwt, invalidateSynaluxJwt } from "../utils/synaluxJwt.js";
 import { resolvePortalBaseUrl, usablePortalKey } from "../utils/synaluxSearch.js";
 import { isSynaluxSignedOut, setSynaluxSignedOut } from "../utils/synaluxCredentialState.js";
@@ -31,6 +32,7 @@ const defaultDeps = {
     getJwt: getSynaluxJwt,
     invalidateJwt: invalidateSynaluxJwt,
     invalidateEntitlements,
+    clearInferencePolicies,
     closeStorage: async () => (await import("../storage/index.js")).closeStorage(),
     resolvePortalBaseUrl,
     usablePortalKey,
@@ -139,6 +141,7 @@ function normalizeAccount(value, origin) {
         role_key: typeof raw.role_key === "string" ? raw.role_key.slice(0, 80) : null,
         plan: plan,
         subscription_plan: subscriptionPlan,
+        prism_plan: raw.prism_plan === "pro" || raw.prism_plan === "team" ? raw.prism_plan : null,
         plan_source: raw.plan_source === "managed" ? "managed" : "stripe",
         billing_status: billingStatus,
         trial_ends_at: billingStatus === "trialing" || billingStatus === "sync_pending" ? trialEndsAt : null,
@@ -245,6 +248,7 @@ export async function connectDashboardAccount(code, overrides = {}) {
         process.env.PRISM_SYNALUX_API_KEY = token;
         deps.invalidateJwt();
         deps.invalidateEntitlements();
+        deps.clearInferencePolicies();
         await deps.closeStorage();
         return accountWithJwt(deps, origin);
     });
@@ -298,6 +302,7 @@ export async function signOutDashboardAccount(overrides = {}) {
         delete process.env.PRISM_SYNALUX_API_KEY;
         deps.invalidateJwt();
         deps.invalidateEntitlements();
+        deps.clearInferencePolicies();
         await deps.setSetting("PRISM_SYNALUX_SIGNED_OUT", "true");
         await deps.setSetting("PRISM_SYNALUX_API_KEY", "");
         await deps.closeStorage();

package/dist/dashboard/ui.js

@@ -1428,7 +1428,7 @@ export function renderDashboardHTML(version) {
       </div>
 
       <div style="margin-top:1.25rem;padding:0.85rem;border-radius:var(--radius-sm);background:rgba(16,185,129,0.08);border:1px solid rgba(16,185,129,0.2);font-size:0.78rem;color:var(--text-secondary)">
-        ✅ <strong>All 6 enforcement layers are active.</strong> Ethics enforcement is NOT tier-gated — applies equally to Free, Standard, Advanced, and Enterprise.
+        ✅ <strong>All 6 enforcement layers are active.</strong> Ethics enforcement is NOT tier-gated — applies equally to Free, the free account, Pro and Team.
         <a href="/ethics" style="color:var(--accent-purple);margin-left:0.5rem">View full policy →</a>
       </div>
     </div>
@@ -2109,13 +2109,22 @@ var _searchBoost = document.getElementById('searchContextBoost');
 if (_searchBoost)
     _searchBoost.addEventListener('change', performSearch);
 var currentAccount = null;
+// Prism's own Pro and Team arrive as the tier they include (plan) plus
+// prism_plan; show their names when the tier is theirs.
+function accountDisplayPlan(account) {
+    if (account && account.prism_plan === 'pro' && account.plan === 'standard') return 'pro';
+    if (account && account.prism_plan === 'team' && account.plan === 'advanced') return 'team';
+    return account ? account.plan : 'free';
+}
 function accountPlanLabel(plan) {
-    var labels = { free: 'Free', standard: 'Standard', advanced: 'Advanced', enterprise: 'Enterprise' };
+    var labels = { free: 'Free', pro: 'Pro', team: 'Team', standard: 'Standard', advanced: 'Advanced', enterprise: 'Enterprise' };
     return labels[plan] || 'Free';
 }
 function accountPlanSummary(plan) {
     var summaries = {
         free: 'Local memory, local models, and the Mind Palace dashboard remain available without an account.',
+        pro: 'Prism Pro is active: cloud answers, answer confirmation, sync, and the full skill library.',
+        team: 'Prism Team is active for this seat: cloud answers, answer confirmation, sync, and team skills.',
         standard: 'Cloud sync and paid Prism services are active for this account.',
         advanced: 'Advanced cloud capacity and team features are active for this account.',
         enterprise: 'Enterprise capacity, governance, and support are active for this account.'
@@ -2141,13 +2150,13 @@ function accountBillingBadge(account) {
 function accountSubscriptionSummary(account) {
     if (account.billing_status === 'trialing') {
         var end = accountTrialEndLabel(account.trial_ends_at);
-        return accountPlanLabel(account.plan) + ' trial is active' + (end ? ' through ' + end : '') +
+        return accountPlanLabel(accountDisplayPlan(account)) + ' trial is active' + (end ? ' through ' + end : '') +
             '. Add payment details before the trial ends to continue; otherwise it cancels automatically.';
     }
     if (['past_due', 'unpaid', 'incomplete', 'paused'].indexOf(account.billing_status) >= 0) {
         var billedPlan = account.subscription_plan
             ? accountPlanLabel(account.subscription_plan) + ' plan'
-            : account.plan === 'free' ? 'paid subscription' : accountPlanLabel(account.plan) + ' plan';
+            : account.plan === 'free' ? 'paid subscription' : accountPlanLabel(accountDisplayPlan(account)) + ' plan';
         return 'Your ' + billedPlan + ' needs billing attention. Update payment details to restore or keep cloud features active.';
     }
     if (account.billing_status === 'sync_pending') {
@@ -2155,7 +2164,7 @@ function accountSubscriptionSummary(account) {
         var syncTrialEnd = accountTrialEndLabel(account.trial_ends_at);
         return 'Stripe confirms your ' + syncingPlan + ' subscription' +
             (syncTrialEnd ? ' with a trial through ' + syncTrialEnd : '') +
-            '. Synalux is updating access; current access remains ' + accountPlanLabel(account.plan) + '. Retry shortly or manage the subscription.';
+            '. Synalux is updating access; current access remains ' + accountPlanLabel(accountDisplayPlan(account)) + '. Retry shortly or manage the subscription.';
     }
     if (account.billing_status === 'canceled' || account.billing_status === 'incomplete_expired') {
         return 'This paid subscription is no longer active. Local Prism Free remains available.';
@@ -2163,7 +2172,7 @@ function accountSubscriptionSummary(account) {
     if (account.billing_status === 'unknown') {
         return 'Prism could not verify the current trial or payment status. Open billing to review the subscription; local features remain available.';
     }
-    return accountPlanSummary(account.plan);
+    return accountPlanSummary(accountDisplayPlan(account));
 }
 function renderPlanLadder(plan) {
     var tiers = [
@@ -2199,7 +2208,7 @@ function renderAccountChip(account, error) {
             ? planStatus
             : account.plan === 'free' && planStatus === 'Payment due'
                 ? planStatus
-                : accountPlanLabel(account.plan) + planStatus;
+                : accountPlanLabel(accountDisplayPlan(account)) + planStatus;
         chip.innerHTML = '<span class="role-icon">🤖</span><span class="identity-label">' + escapeHtml(account.name || 'Synalux user') + '</span>' +
             '<span class="plan-mini' + planStatusClass + '">' + escapeHtml(chipPlanLabel) + '</span>';
     }
@@ -2249,7 +2258,7 @@ function renderAccountPanel(account, error) {
         account.billing && account.billing.action === 'included' ? 'View plans' : 'Start 14-day trial';
     managedBadge = account.plan_source === 'managed' ? '<span class="account-badge managed">Managed</span>' : '';
     panel.innerHTML = '<div class="account-card"><div class="account-card-head"><div>' +
-        '<div class="account-badges"><span class="account-badge">' + escapeHtml(accountPlanLabel(account.plan)) + '</span>' +
+        '<div class="account-badges"><span class="account-badge">' + escapeHtml(accountPlanLabel(accountDisplayPlan(account))) + '</span>' +
         '<span class="account-badge role">' + escapeHtml(role) + '</span>' + managedBadge + accountBillingBadge(account) + '</div>' +
         '<div class="account-name">' + escapeHtml(account.name || 'Synalux user') + '</div>' +
         '<div class="account-summary">' + escapeHtml(accountSubscriptionSummary(account)) + '</div></div></div>' +

package/dist/server.js

@@ -60,7 +60,7 @@ ListResourcesRequestSchema, ListResourceTemplatesRequestSchema, ReadResourceRequ
 // Claude Desktop that the attached resource has changed.
 // Without this, the paperclipped context becomes stale.
 SubscribeRequestSchema, UnsubscribeRequestSchema, } from "@modelcontextprotocol/sdk/types.js";
-import { SERVER_CONFIG, SESSION_MEMORY_ENABLED, PRISM_USER_ID, PRISM_ENABLE_HIVEMIND, WATCHDOG_INTERVAL_MS, WATCHDOG_STALE_MIN, WATCHDOG_FROZEN_MIN, WATCHDOG_OFFLINE_MIN, WATCHDOG_LOOP_THRESHOLD, PRISM_SCHEDULER_ENABLED, PRISM_SCHEDULER_INTERVAL_MS, PRISM_HDC_ENABLED, PRISM_TASK_ROUTER_ENABLED_ENV, PRISM_DARK_FACTORY_ENABLED, } from "./config.js";
+import { SERVER_CONFIG, SESSION_MEMORY_ENABLED, PRISM_USER_ID, PRISM_ENABLE_HIVEMIND, WATCHDOG_INTERVAL_MS, WATCHDOG_STALE_MIN, WATCHDOG_FROZEN_MIN, WATCHDOG_OFFLINE_MIN, WATCHDOG_LOOP_THRESHOLD, PRISM_SCHEDULER_ENABLED, PRISM_SCHEDULER_INTERVAL_MS, PRISM_HDC_ENABLED, PRISM_TASK_ROUTER_ENABLED_ENV, PRISM_DARK_FACTORY_ENABLED, YDC_API_KEY, } from "./config.js";
 import { startWatchdog, drainAlerts } from "./hivemindWatchdog.js";
 import { startScheduler } from "./backgroundScheduler.js";
 import { startDarkFactoryRunner } from "./darkfactory/runner.js";
@@ -91,7 +91,7 @@ import { triggerSkillManifestSync } from "./skillManifestSync.js";
 import { EVIDENCE_WORKFLOW_POLICY_TEXT } from "./evidenceWorkflowPolicy.js";
 import { LOCAL_FIRST_POLICY_TEXT } from "./localFirstPolicy.js";
 // ─── Import Tool Definitions (schemas) and Handlers (implementations) ─────
-import { WEB_SEARCH_TOOL, BRAVE_WEB_SEARCH_CODE_MODE_TOOL, LOCAL_SEARCH_TOOL, BRAVE_LOCAL_SEARCH_CODE_MODE_TOOL, CODE_MODE_TRANSFORM_TOOL, BRAVE_ANSWERS_TOOL, RESEARCH_PAPER_ANALYSIS_TOOL, webSearchHandler, braveWebSearchCodeModeHandler, localSearchHandler, braveLocalSearchCodeModeHandler, codeModeTransformHandler, braveAnswersHandler, researchPaperAnalysisHandler, } from "./tools/index.js";
+import { WEB_SEARCH_TOOL, BRAVE_WEB_SEARCH_CODE_MODE_TOOL, LOCAL_SEARCH_TOOL, BRAVE_LOCAL_SEARCH_CODE_MODE_TOOL, CODE_MODE_TRANSFORM_TOOL, BRAVE_ANSWERS_TOOL, RESEARCH_PAPER_ANALYSIS_TOOL, YOUCOM_WEB_SEARCH_TOOL, webSearchHandler, braveWebSearchCodeModeHandler, localSearchHandler, braveLocalSearchCodeModeHandler, codeModeTransformHandler, braveAnswersHandler, researchPaperAnalysisHandler, youcomWebSearchHandler, } from "./tools/index.js";
 // Session memory tools — only used if Supabase is configured
 import { SESSION_SAVE_LEDGER_TOOL, SESSION_SAVE_HANDOFF_TOOL, SESSION_LOAD_CONTEXT_TOOL, SESSION_BOOTSTRAP_TOOL, SESSION_ROUTE_PROMPT_TOOL, KNOWLEDGE_SEARCH_TOOL, KNOWLEDGE_FORGET_TOOL, 
 // ─── v0.4.0: New tool definitions (Enhancements #2 and #4) ───
@@ -180,6 +180,8 @@ const BASE_TOOLS = [
     BRAVE_ANSWERS_TOOL, // brave_answers — AI-grounded answers
     RESEARCH_PAPER_ANALYSIS_TOOL, // gemini_research_paper_analysis — paper analysis
     PRISM_INFER_TOOL, // prism_infer — local-first inference (token-saving cascade)
+    // youcom_web_search — optional You.com web search, gated on YDC_API_KEY
+    ...(YDC_API_KEY ? [YOUCOM_WEB_SEARCH_TOOL] : []),
 ];
 // Hook-free first-turn startup is owned by session_bootstrap. Its static tool
 // description is surfaced by native MCP hosts before any tool call and keeps
@@ -748,6 +750,10 @@ export function createServer() {
                     case "prism_infer":
                         result = await prismInferHandler(args);
                         break;
+                    // ── Optional You.com Web Search (only when YDC_API_KEY is set) ──
+                    case "youcom_web_search":
+                        result = await youcomWebSearchHandler(args);
+                        break;
                     // ── Session Memory Tools (only callable when Supabase is configured) ──
                     // REVIEWER NOTE: Even though these tools won't appear in the
                     // tool list without Supabase, we still guard each handler call

package/dist/storage/inferMetricsLedger.js

@@ -42,8 +42,8 @@ const LEDGER_UNAVAILABLE_ERROR = "Inference metrics ledger is unavailable";
 const INSERT_METRIC_SQL = `INSERT OR IGNORE INTO infer_metrics
     (ts, caller, mode, backend, model, used_cloud, gate_outcome,
      refusal_reason, prompt_tokens, completion_tokens, latency_ms, ram_free_mb,
-     source_event_id, history_turns, refusal_layer)
-    VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)`;
+     source_event_id, history_turns, refusal_layer, layer1_second_read)
+    VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)`;
 function closeClient(context) {
     const activeClient = client;
     client = null;
@@ -85,7 +85,8 @@ function ensureTable() {
                     ram_free_mb INTEGER,
                     source_event_id TEXT,
                     history_turns INTEGER,
-                    refusal_layer TEXT
+                    refusal_layer TEXT,
+                    layer1_second_read TEXT
                 )`);
             // Existing ledgers predate external panel-spool ingestion. SQLite has
             // no ADD COLUMN IF NOT EXISTS, so use the repository's established
@@ -94,6 +95,7 @@ function ensureTable() {
                 "source_event_id TEXT",
                 "history_turns INTEGER",
                 "refusal_layer TEXT",
+                "layer1_second_read TEXT",
             ]) {
                 try {
                     await client.execute(`ALTER TABLE infer_metrics ADD COLUMN ${column}`);
@@ -158,6 +160,7 @@ function metricArgs(row) {
         row.completion_tokens ?? null, row.latency_ms ?? null,
         row.ram_free_mb ?? null, row.source_event_id ?? null,
         row.history_turns ?? null, row.refusal_layer ?? null,
+        row.layer1_second_read ?? null,
     ];
 }
 /** Aggregate all persisted rows (optionally since a timestamp). */
@@ -284,6 +287,32 @@ export async function queryLocalSavings(sinceTs) {
                   GROUP BY COALESCE(model, backend)`,
             args: whereArgs,
         });
+        const followWhere = `WHERE history_turns > 0${sinceTs != null ? " AND ts >= ?" : ""}`;
+        const PLAN_REFUSAL = `COALESCE(refusal_reason, '') IN ('multi_turn_not_in_plan', 'history_over_plan_cap')`;
+        // A 9b model: '9b' right after a non-digit, so a future '19b' is not counted.
+        const NINE_B = `(LOWER(COALESCE(model, backend)) GLOB '*[^0-9]9b*' OR LOWER(COALESCE(model, backend)) GLOB '9b*')`;
+        const follow = await client.execute({
+            sql: `SELECT
+                SUM(CASE WHEN ${SERVED_LOCAL} THEN 1 ELSE 0 END) AS served,
+                SUM(CASE WHEN ${SERVED_LOCAL} AND ${NINE_B} THEN 1 ELSE 0 END) AS served_9b,
+                SUM(CASE WHEN used_cloud = 1 THEN 1 ELSE 0 END) AS cloud,
+                SUM(CASE WHEN used_cloud = 0 AND NOT (${SERVED_LOCAL}) AND NOT (${PLAN_REFUSAL}) THEN 1 ELSE 0 END) AS refused,
+                SUM(CASE WHEN used_cloud = 0 AND NOT (${SERVED_LOCAL}) AND ${PLAN_REFUSAL} THEN 1 ELSE 0 END) AS refused_by_plan
+                  FROM infer_metrics ${followWhere}`,
+            args: whereArgs,
+        });
+        const followLayers = await client.execute({
+            sql: `SELECT COALESCE(refusal_layer, 'unrecorded') AS layer, COUNT(*) AS n
+                  FROM infer_metrics ${followWhere} AND used_cloud = 0 AND NOT (${SERVED_LOCAL})
+                        AND NOT (${PLAN_REFUSAL})
+                  GROUP BY COALESCE(refusal_layer, 'unrecorded')`,
+            args: whereArgs,
+        });
+        const f = follow.rows[0];
+        const refused_by_layer = {};
+        for (const row of followLayers.rows) {
+            refused_by_layer[String(row.layer)] = Number(row.n ?? 0);
+        }
         const r = agg.rows[0];
         const by_model = {};
         for (const row of byM.rows) {
@@ -310,6 +339,14 @@ export async function queryLocalSavings(sinceTs) {
             first_ts: r.first_ts == null ? null : Number(r.first_ts),
             last_ts: r.last_ts == null ? null : Number(r.last_ts),
             by_model,
+            followups: {
+                served_local: Number(f.served ?? 0),
+                served_local_9b: Number(f.served_9b ?? 0),
+                refused: Number(f.refused ?? 0),
+                refused_by_plan: Number(f.refused_by_plan ?? 0),
+                cloud: Number(f.cloud ?? 0),
+                refused_by_layer,
+            },
         };
     }
     catch (e) {