@synapsor/runner 1.6.3 → 1.6.5

package/docs/security-boundary.md

@@ -42,6 +42,36 @@ The model does not receive:
 - MCP endpoint tokens, JWTs, refresh tokens, client secrets, or TLS keys;
 - trusted tenant or principal authority as ordinary model arguments.
 
+## Optional Workbench model client
+
+Workbench Ask is a local client of that same reviewed surface, not another
+policy engine. It is available only from authenticated loopback Workbench in an
+explicit development/staging profile. Its provider may request only the exact
+active named capabilities plus valid local authoring tools. The request is
+dispatched through the official MCP SDK and the existing typed-plan/runtime
+validators; there is no parallel query or mutation executor.
+
+The operator chooses provider, model, endpoint, and credential source and
+acknowledges direct egress for the current authority digest. No model output can
+change those values. Provider credentials and bounded conversation state stay
+in local server memory and are not written to the project, ledger, browser
+storage, logs, evidence, receipts, or replay. Reviewed tool results may be sent
+directly to the provider after consent; Synapsor does not relay them.
+
+Official remote providers use fixed HTTPS origins. Custom remote endpoints
+require HTTPS; plaintext is restricted to loopback. Redirects are refused,
+resolved destinations are checked and pinned on every connection, and
+private/link-local/metadata/special remote addresses are blocked. Unknown
+tools, malformed arguments/responses, authority drift, oversized data, time or
+tool-loop exhaustion, and reported source mutation fail closed.
+
+Provider prose remains untrusted. A write tool can create only the normal inert
+proposal, and the Workbench directs the human to the separate operator path.
+Ask never receives activation, Protect, approval, apply, worker, notification,
+attention, or reconciliation tools. It is absent from production,
+unknown-profile, shared, remote, and non-loopback surfaces. See [Workbench Ask
+With Your Model](workbench-ask.md).
+
 ## Network transport and identity
 
 Local stdio is the preferred one-client path. The MCP client launches Runner and

package/docs/troubleshooting-first-run.md

@@ -3,26 +3,26 @@
 Run the friendly doctor first:
 
 ```bash
-npx -y -p @synapsor/runner synapsor-runner doctor --first-run
+npx -y @synapsor/runner doctor --first-run
 ```
 
 Use JSON for automation:
 
 ```bash
-npx -y -p @synapsor/runner synapsor-runner doctor --first-run --json
+npx -y @synapsor/runner doctor --first-run --json
 ```
 
 ## Guided Recovery Contract
 
-Runner 1.6.3 failures should tell you what failed, why the boundary stopped,
+Runner 1.6.4 failures should tell you what failed, why the boundary stopped,
 what state remains, and one next action. Do not delete the project or add
 `--force` merely to recover.
 
 | Failure | State preserved | One next action |
 | --- | --- | --- |
-| Database connection or metadata inspection failed | Existing project/review files and source rows | Fix the exported URL/network without printing the credential, then rerun `npx -y @synapsor/runner@latest start --from-env DATABASE_URL`. |
+| Database connection or metadata inspection failed | Existing project/review files and source rows | Fix the URL/network without printing the credential, then rerun `npx -y @synapsor/runner start`. |
 | Read role is writable, owner, superuser, `BYPASSRLS`, or unverifiable | Disabled metadata draft; no source-row Explore | Supply a verifiably SELECT-only non-owner staging role, then rerun the same `start` command. |
-| Schema or project choice is ambiguous | Existing files; no authority activation | Rerun with the exact reviewed schema, for example `npx -y @synapsor/runner@latest start --from-env DATABASE_URL --schema public`. |
+| Schema or project choice is ambiguous | Existing files; no authority activation | Rerun with the exact reviewed schema, for example `npx -y @synapsor/runner start --schema public`. |
 | Tenant or principal scope is unresolved | Conservative blocked resource decisions | Open the same Workbench URL and resolve the highlighted scope exception. |
 | Sensitive field remains unresolved | Field stays kept out; active tools unchanged | Open Workbench **Exceptions** and record one reviewed field decision. |
 | Row identifier is missing/composite/ambiguous | Resource remains blocked | Select a source-proven single-column primary/unique identity or keep the resource blocked. |
@@ -33,6 +33,9 @@ what state remains, and one next action. Do not delete the project or add
 | Config mode is missing/invalid | Config and source database are unchanged | Set `mode` to `read_only`, `shadow`, `review`, or `cloud`, then rerun `synapsor-runner config validate --json`. |
 | Config contains an unknown field | Config and source database are unchanged | Remove or correct the reported JSON path, then rerun `synapsor-runner config validate --json`. |
 | Workbench port is occupied | Review files, ledger, and source database | Rerun `synapsor-runner ui --open`; the default selects a free loopback port. |
+| Workbench Ask is missing | Reviewed project remains usable through no-model Workbench/CLI/MCP | Use authenticated loopback Workbench with an explicit development/staging profile and at least one active reviewed tool. |
+| Ask provider/key fails | Boundary, no-model composer, and external MCP remain usable | Check the exported key name, provider/model access, and local DNS/TLS; then retry without changing authority. |
+| Ask says authority changed | Previous consent is invalid; active reviewed tools remain unchanged | Inspect the new tool/digest summary and acknowledge direct egress again. |
 | Writeback setup fails | Config and reviewed plan; transactional setup rolls back | Rerun `synapsor-runner writeback setup --profile staging --json` and review the reported prerequisite. |
 | Writer role/setup URL is missing | No DDL or grant was applied | Rerun the preview with `--writer-role <role> --setup-url-env <ADMIN_URL_ENV>`. |
 | No supported write candidate exists | Read boundary remains active; no write authority | Use **Add a safe action** on a writable base table with a proven identity/version field, or retain read-only mode. |
@@ -49,9 +52,10 @@ outcome; inspect the durable receipt/reconciliation state rather than guessing.
 
 ## Fresh Start Did Not Enter Auto Boundary
 
-Auto Boundary is the default only for a fresh interactive `start --from-env`
-with no existing config, selector, answers file, machine-output flag, or other
-automation input. This preserves every established 1.x route.
+Auto Boundary is the default only for a fresh interactive `start` with no
+existing config, selector, answers file, machine-output flag, or other
+automation input. An exported `DATABASE_URL` is implied; explicit connection
+and schema flags still win. This preserves every established 1.x route.
 
 Check the generated state:
 
@@ -80,9 +84,10 @@ Check:
 ```bash
 synapsor-runner boundary status --json
 synapsor-runner boundary diff --json
-synapsor-runner mcp status cursor --project
+synapsor-runner mcp status claude-code --project
 ```
 
+Use `cursor` or `vscode` instead when that is the selected client.
 The exact boundary digest must be active, the profile must explicitly be
 `development` or `staging`, the generation lock and compiler/Spec versions
 must be current, and the inspected database role must still be SELECT-only,
@@ -93,12 +98,48 @@ A write-capable, owner, superuser, `BYPASSRLS`, or unverifiable credential may
 still inspect metadata with a warning. It cannot enable source-row Explore.
 Use a dedicated staging reader instead of weakening this check.
 
-## Cursor Has Production Tools Instead Of Authoring Tools
+## Workbench Ask Is Missing Or Refused
+
+Ask is optional and follows a stricter local profile boundary. It appears only
+in authenticated loopback Workbench for explicit `development` or `staging`
+projects with at least one reviewed tool. Production, unknown, shared, remote,
+and non-loopback surfaces intentionally omit it.
+
+Normal Runner and MCP execution do not automatically source `.env`. Fresh
+interactive `start` may read a regular project `.env` only after explicit
+consent and keeps the selected database URL in that process. For the optional
+model provider, export the key in the same shell that launches Workbench, then
+select **Read an environment variable** and enter only the variable name. A
+session-only masked paste is also supported.
+
+Provider errors are redacted and do not disable the no-model composer:
+
+- `ASK_KEY_REQUIRED`: export the selected key or use the masked session paste;
+- `ASK_AUTHORITY_CHANGED`: review the current tools/digest and acknowledge
+  direct egress again;
+- `ASK_PROVIDER_UNAVAILABLE`: verify provider availability, key/model access,
+  and local DNS/TLS, then retry;
+- `ASK_PROVIDER_REDIRECT_REFUSED`: use the final fixed endpoint; Runner never
+  forwards credentials across redirects;
+- `ASK_PROVIDER_DESTINATION_REFUSED`: the host resolved to a private, special,
+  metadata, or otherwise disallowed remote address;
+- `ASK_REMOTE_HTTPS_REQUIRED`: remote custom endpoints require HTTPS; HTTP is
+  accepted only on loopback;
+- `ASK_TOOL_REQUIRED`: the provider answered from prose without using a
+  reviewed Synapsor tool;
+- `ASK_UNKNOWN_TOOL` or `ASK_OPERATOR_TOOL_REFUSED`: the provider requested
+  authority outside the displayed tool surface.
+
+Select **Clear** after a session to cancel active work and discard in-memory
+provider configuration/history. See [Workbench Ask With Your
+Model](workbench-ask.md).
+
+## A Project Client Has Production Tools Instead Of Authoring Tools
 
 Install the managed local authoring entry only after boundary activation:
 
 ```bash
-synapsor-runner mcp install cursor \
+synapsor-runner mcp install claude-code \
   --project \
   --authoring \
   --project-root . \
@@ -111,13 +152,15 @@ production config. The protected named capability remains available while
 Explore disappears:
 
 ```bash
-synapsor-runner mcp install cursor \
+synapsor-runner mcp install claude-code \
   --project \
   --config ./synapsor.runner.json \
   --store ./.synapsor/local.db \
   --yes
 ```
 
+Use `cursor` or `vscode` instead of `claude-code` for those clients.
+
 ## Boundary Became Stale After A Schema Or Grant Change
 
 Generated authority is bound to the schema, role, grants, ownership, RLS
@@ -164,19 +207,20 @@ Open the secured Workbench, run the real source-unchanged staging Data PR
 preview, review the complete digest, and activate it there. There is
 intentionally no activation CLI command to hand to a coding agent.
 
-## Activated Tool Does Not Appear In Cursor
+## Activated Tool Does Not Appear In The MCP Client
 
 Some MCP hosts do not refresh `tools/list` for a running stdio session. First
 confirm that Runner's active tool surface changed:
 
 ```bash
 synapsor-runner action status --json
-synapsor-runner mcp status cursor --project --check-launch
+synapsor-runner mcp status claude-code --project --check-launch
 ```
 
+Use `cursor` or `vscode` instead of `claude-code` for those clients.
 Then reconnect or restart the project MCP server as directed by the Workbench.
 Do not work around a stale host session by adding approval, apply, activation,
-credentials, tenant values, or raw SQL to the Cursor configuration.
+credentials, tenant values, or raw SQL to the client configuration.
 
 ## Safe Action Validation Reports Review Placeholders
 
@@ -327,13 +371,13 @@ Own-database MCP setup needs a reviewed config before serving tools.
 Fix:
 
 ```bash
-npx -y -p @synapsor/runner synapsor-runner init --from-env DATABASE_URL --mode review --wizard
+npx -y @synapsor/runner init --from-env DATABASE_URL --mode review --wizard
 ```
 
 Or pass an example config:
 
 ```bash
-npx -y -p @synapsor/runner synapsor-runner tools preview --config ./examples/mcp-postgres-billing/synapsor.runner.json --store ./.synapsor/local.db
+npx -y @synapsor/runner tools preview --config ./examples/mcp-postgres-billing/synapsor.runner.json --store ./.synapsor/local.db
 ```
 
 ## SQLite Store Missing
@@ -379,7 +423,7 @@ Fix:
 
 ```bash
 export SYNAPSOR_DATABASE_READ_URL="<read-only-url>"
-npx -y -p @synapsor/runner synapsor-runner doctor --config synapsor.runner.json
+npx -y @synapsor/runner doctor --config synapsor.runner.json
 ```
 
 ## Read/Write Credential Split Failed
@@ -491,7 +535,7 @@ Fix:
 Regenerate the snippet:
 
 ```bash
-npx -y -p @synapsor/runner synapsor-runner mcp config claude-desktop \
+synapsor-runner mcp config claude-desktop \
   --absolute-paths \
   --config ./synapsor.runner.json \
   --store ./.synapsor/local.db

package/docs/use-your-own-database.md

@@ -10,9 +10,11 @@ model.
 Short path:
 
 ```bash
-export DATABASE_URL="postgresql://readonly_user:password@host:5432/app?sslmode=require"
-npx -y -p @synapsor/runner synapsor-runner start --from-env DATABASE_URL --schema public
+npx -y @synapsor/runner start
 ```
 
-Runner stores environment-variable names in `synapsor.runner.json`, not database
-URLs. Keep credentials in your shell, process manager, or secret manager.
+Paste a read-only URL into the hidden prompt, explicitly approve a regular
+project `.env` file for this process, or export `DATABASE_URL`. PostgreSQL
+defaults to `public`; explicit `--from-env`, `--schema`, and `--project-root`
+flags remain available and always win. Runner stores environment-variable names
+in `synapsor.runner.json`, not database URLs.

package/docs/workbench-ask.md

@@ -0,0 +1,279 @@
+# Workbench Ask With Your Model
+
+Workbench Ask is an optional local client for the reviewed Synapsor tools that
+already exist. It lets a developer ask a plain-language question through
+OpenAI, Anthropic, or an explicitly configured OpenAI-compatible endpoint
+without installing an external MCP host first.
+
+Ask does not generate or activate authority. It does not replace the no-model
+Workbench composer, which remains the default onboarding path and requires no
+model API key.
+
+```text
+human activates a reviewed boundary or named capability
+  -> Workbench lists that exact model-facing tool surface
+  -> human selects a provider and acknowledges direct data egress
+  -> provider may request one of those reviewed tools
+  -> Runner executes the request through the normal MCP/runtime path
+  -> provider receives the bounded reviewed result
+  -> Workbench renders provider prose as untrusted output
+```
+
+The model cannot activate a boundary, Protect a query, approve or apply a
+proposal, control a worker, acknowledge attention, reconcile an outcome, or
+change provider configuration.
+
+## Availability
+
+Ask appears only when all of these conditions hold:
+
+- Workbench is authenticated and bound to loopback;
+- the deployment profile is explicitly `development` or `staging`;
+- at least one reviewed model-facing tool is available;
+- generated authority and its generation lock are current when applicable.
+
+It is absent from production, missing/unknown profiles, non-loopback
+Workbench, shared remote HTTP, and model-facing MCP configuration. An MCP
+request cannot enable Ask or select its provider.
+
+While Scoped Explore is active, Ask may use:
+
+```text
+app.describe_data
+app.explore_data
+```
+
+After Explore is disabled, Ask may use only the activated named capabilities
+in `synapsor.runner.json`, such as:
+
+```text
+analytics.weekly_revenue_by_store
+retail.propose_order_fulfillment
+```
+
+The panel displays the exact tool names and an Ask authority digest. That digest
+binds the tool surface, runtime config, deployment profile, and active
+exploration digest when one exists. A material authority change invalidates
+egress consent before another provider call.
+
+## Start Without A Model
+
+Begin with the normal one-command staging path:
+
+```bash
+export SYNAPSOR_TENANT_ID='<reviewed tenant>'
+export SYNAPSOR_PRINCIPAL='<reviewed principal>'
+npx -y @synapsor/runner start
+```
+
+Paste the SELECT-only database URL into Runner's hidden terminal prompt, or
+explicitly consent to loading only `DATABASE_URL` from a recognized project
+environment file. For an intentionally exported automation path, use
+`start --from-env DATABASE_URL`.
+
+Review and activate the boundary, then run the first safe read or aggregate in
+Workbench. At this point the complete no-model path is already working.
+
+Open **Analyze**, then **Ask with your model**. Workbench shows:
+
+- the reviewed tools the provider may request;
+- the current boundary/authority digest;
+- provider, model, origin, and credential source;
+- that approved visible results may go directly to the provider;
+- that Synapsor does not relay the request;
+- that the source database is unchanged by a read or proposal call.
+
+## Provider Credentials
+
+Workbench Ask does not load provider keys from `.env` files. Instant database
+onboarding may separately read only a selected `DATABASE_URL` after explicit
+human consent. Export a provider key in the same shell that starts Runner:
+
+```bash
+export OPENAI_API_KEY='<provider key>'
+npx -y @synapsor/runner start
+```
+
+Provide the database URL through the same hidden/consented path above. If
+Runner is already running, use the masked Workbench key field instead of
+restarting only to add a provider key.
+
+In Workbench select:
+
+```text
+Provider: OpenAI
+Credential: Read an environment variable
+Environment variable name: OPENAI_API_KEY
+```
+
+The browser sends only the environment variable name. The value stays in the
+local Runner process and is not returned to the browser.
+
+Alternatively, paste a key into the masked Workbench field. Runner holds that
+value only in server memory for the current local session and clears the field
+after configuration. Do not paste credentials into an agent chat.
+
+Keys and authorization headers are not written to:
+
+- generated DSL, JSON, locks, or config;
+- the SQLite/shared ledger, evidence, receipts, or replay;
+- browser local/session storage;
+- logs, screenshots, telemetry, or command-line arguments.
+
+Select **Clear** to cancel an active request, remove in-memory provider
+configuration, and discard bounded conversation history. Stopping Workbench
+also clears it.
+
+## Direct Egress
+
+Before the first call, Workbench requires this provider-specific decision:
+
+```text
+Your reviewed visible data may be sent directly to this provider using your
+credential. Synapsor does not receive it. Fields kept out by the active
+boundary are unavailable to this model.
+```
+
+Consent is bound to the provider, model, endpoint origin, and current Ask
+authority digest. Changing any of those requires a new acknowledgement.
+Consent does not activate or widen database authority.
+
+The question, reviewed tool definitions, and bounded tool results go directly
+from local Runner to the selected provider. A tool result can contain reviewed
+source values, so use a provider and data-handling policy appropriate for that
+data. Kept-out fields are unavailable to selection, filtering, grouping,
+sorting, joins, or returned results.
+
+Runner does not persist the conversation, provider response, or returned tool
+rows by default. Existing normalized query audit still records the boundary
+digest, keyed plan/literal fingerprints, suppression decision, timing, and
+result size without storing result values or trusted tenant/principal values.
+An operator-created screenshot or external provider may retain data under its
+own policy; Runner cannot erase those external copies.
+
+## Authority And Proposal Behavior
+
+Provider tool calls use the official MCP SDK and the same canonical runtime as
+an external client:
+
+- typed argument validation;
+- trusted tenant/principal injection outside model arguments;
+- read-only transaction enforcement for Scoped Explore;
+- field, relationship, suppression, extraction, and differencing limits;
+- proposal construction and exact active contract digest;
+- the same error envelope and source-change state.
+
+Provider prose, tool names, arguments, and response metadata remain untrusted.
+An unknown or operator-plane tool is refused. A malformed or oversized
+request, repeated tool loop, changed authority, timeout, or provider failure
+stops safely.
+
+A reviewed write tool may create an inert proposal:
+
+```text
+Proposal only
+The source database did not change.
+The model cannot approve or apply this proposal.
+```
+
+The human continues through the separate Workbench/CLI operator workflow.
+Ask never receives approval or commit tools.
+
+## Endpoint Security
+
+Official providers use fixed endpoints:
+
+```text
+OpenAI:    https://api.openai.com/v1/chat/completions
+Anthropic: https://api.anthropic.com/v1/messages
+```
+
+Use **Custom OpenAI-compatible** for another endpoint. Remote custom endpoints
+must use HTTPS. Plain HTTP is accepted only for an explicit loopback endpoint,
+for example:
+
+```text
+http://127.0.0.1:11434/v1
+```
+
+Runner:
+
+- rejects URL credentials, fragments, and query parameters;
+- disables redirects and never forwards a key across one;
+- resolves and validates DNS for every connection;
+- pins the connection to the validated address while retaining TLS hostname
+  verification;
+- blocks private, link-local, multicast, unspecified, metadata, and other
+  special destinations for remote mode;
+- permits custom local mode only on loopback;
+- bounds request/response bytes, tool schemas/results, tool calls, iterations,
+  history, time, and reported tokens.
+
+No URL, model, header, or destination comes from model output.
+
+## Fixed Session Bounds
+
+The current release enforces:
+
+| Item | Bound |
+| --- | ---: |
+| Question | 4,000 characters |
+| Model identifier | 128 characters |
+| Provider request | 256 KiB |
+| Provider response | 1 MiB |
+| Tool schema surface | 128 KiB and 64 tools |
+| One tool result | 128 KiB |
+| Tool calls | 4 per provider response, 8 per turn |
+| Tool-loop iterations | 6 |
+| Conversation history | 4 completed turns, 16 KiB |
+| Final answer | 16 KiB |
+| Provider request timeout | 30 seconds |
+| Reported session token usage | 50,000 tokens |
+
+One Workbench Ask session runs one request at a time. Runner does not
+automatically retry provider calls in this release. A developer may retry a
+known safe failure from the UI; every retry begins with current authority
+validation. Token accounting depends on usage reported by the provider and is
+not a monetary spend guarantee.
+
+## Tested Provider Matrix
+
+Status as of the prepared 1.6.4 source:
+
+| Provider surface | Verification | Claim |
+| --- | --- | --- |
+| OpenAI `gpt-5-mini` | Live packed-Workbench run against real local PostgreSQL on 2026-07-25 | Live tested |
+| Anthropic Messages/tool-use protocol | Deterministic mock server, normal/refusal/error paths | Protocol tested; no live Anthropic account run |
+| Custom OpenAI-compatible loopback | Real local HTTP server plus deterministic tool/refusal/proposal paths | Tested against the documented Chat Completions subset |
+| Ollama or another named compatible server | Not installed in the release environment | Unknown until tested; compatibility is not implied by the label |
+
+The live OpenAI run used `app.describe_data` and `app.explore_data`, matched the
+official MCP aggregate result, changed no source rows, and passed exact-key
+artifact scans. See
+`development/runner-1.6.4-community-solar-results.json`.
+
+## Troubleshooting
+
+**Ask is missing:** confirm loopback Workbench, an explicit development/staging
+profile, and at least one active reviewed tool. Production intentionally omits
+Ask.
+
+**Key required:** export the named variable in the process that launches
+Runner, or use the session-only masked paste. Runner does not load `.env`.
+
+**Authority changed:** inspect the new tool/digest summary and acknowledge
+egress again. Do not bypass the check.
+
+**Provider unavailable:** the reviewed boundary and no-model composer remain
+usable. Check provider status, local DNS/TLS, key validity, and model access.
+Runner returns a redacted failure and does not retry automatically.
+
+**Custom endpoint refused:** use HTTPS remotely or an exact loopback HTTP URL.
+Private network and metadata endpoints are intentionally blocked.
+
+**Provider answers without a tool:** Runner refuses to present prose as a
+database answer until at least one reviewed Synapsor tool succeeds or returns a
+reviewed refusal.
+
+**Proposal did not commit:** that is correct. Open the separate operator
+workflow to review, approve, and guardedly apply it.

package/docs/writeback-executors.md

@@ -59,9 +59,9 @@ It does not claim distributed exactly-once. See
 Use the helper commands before enabling direct SQL writeback:
 
 ```bash
-npx -y -p @synapsor/runner synapsor-runner writeback doctor --config ./synapsor.runner.json
-npx -y -p @synapsor/runner synapsor-runner writeback migration --engine postgres
-npx -y -p @synapsor/runner synapsor-runner writeback grants --engine postgres --writer-role app_writer
+npx -y @synapsor/runner writeback doctor --config ./synapsor.runner.json
+npx -y @synapsor/runner writeback migration --engine postgres
+npx -y @synapsor/runner writeback grants --engine postgres --writer-role app_writer
 ```
 
 `writeback doctor --check-db` connects with the configured writer credential and
@@ -106,7 +106,7 @@ literal values.
 Run after approval:
 
 ```bash
-npx -y -p @synapsor/runner synapsor-runner apply \
+npx -y @synapsor/runner apply \
   --proposal wrp_123 \
   --config ./synapsor.runner.json \
   --store ./.synapsor/local.db
@@ -199,13 +199,13 @@ handler receipt.
 Or generate one into your app:
 
 ```bash
-npx -y -p @synapsor/runner synapsor-runner handler template node-fastify \
+npx -y @synapsor/runner handler template node-fastify \
   --output ./synapsor-writeback-handler.mjs
 
-npx -y -p @synapsor/runner synapsor-runner handler template python-fastapi \
+npx -y @synapsor/runner handler template python-fastapi \
   --output ./synapsor_writeback_handler.py
 
-npx -y -p @synapsor/runner synapsor-runner handler template command \
+npx -y @synapsor/runner handler template command \
   --output ./synapsor-command-handler.mjs
 ```
 

package/examples/auto-boundary-churn/README.md

@@ -1,6 +1,6 @@
 # Auto Boundary Churn Fixture
 
-This synthetic staging fixture is the measured Runner 1.6.0 golden path:
+This synthetic staging fixture is the original Runner 1.6.0 focused path:
 PostgreSQL, Next.js, Prisma, Cursor-compatible local MCP, and the local
 Workbench.
 

package/examples/community-solar-clean-room/README.md

@@ -0,0 +1,7 @@
+# Community Solar Clean-Room Fixture
+
+This synthetic PostgreSQL and Prisma application exists only for the Runner
+1.6.4 clean-room adoption gate. It contains no production data or credentials.
+
+The measured journey starts from a copied packed artifact and this fixture in a
+new temporary directory. Synapsor artifacts are not pre-generated.

package/examples/community-solar-clean-room/docker-compose.yml

@@ -0,0 +1,16 @@
+services:
+  postgres:
+    image: postgres:16
+    environment:
+      POSTGRES_DB: community_solar
+      POSTGRES_USER: solar_admin
+      POSTGRES_PASSWORD: solar_admin_password
+    ports:
+      - "127.0.0.1:55464:5432"
+    healthcheck:
+      test: ["CMD-SHELL", "test \"$(psql -U solar_admin -d community_solar -tAc \"SELECT pg_postmaster_start_time() > initialized_at FROM public.synapsor_fixture_ready LIMIT 1\" 2>/dev/null | xargs)\" = t"]
+      interval: 1s
+      timeout: 3s
+      retries: 60
+    volumes:
+      - ./seed/postgres.sql:/docker-entrypoint-initdb.d/001-schema.sql:ro

package/examples/community-solar-clean-room/package.json

@@ -0,0 +1,15 @@
+{
+  "name": "synapsor-community-solar-clean-room",
+  "version": "0.0.0",
+  "private": true,
+  "dependencies": {
+    "@prisma/client": "^6.0.0",
+    "next": "^15.0.0",
+    "react": "^19.0.0",
+    "react-dom": "^19.0.0"
+  },
+  "devDependencies": {
+    "prisma": "^6.0.0",
+    "typescript": "^5.8.0"
+  }
+}