forgecad 0.9.13 → 0.9.14

package/dist/docs-raw/rl-environments.md

@@ -1,62 +1,57 @@
 # ForgeCAD RL Environments
 
 This is the permanent operating guide for ForgeCAD reinforcement-learning and
-AI-lab reconstruction environments. Temporary notes can record experiments, but
-this document is the stable contract for building, running, evaluating, and
-maintaining reconstruction tasks.
+AI-lab reconstruction environments.
 
 ## Product Goal
 
 ForgeCAD RL environments should let an external lab train or evaluate an agent
 that turns reference physical geometry into readable ForgeCAD source code.
 
-The loop is intentionally simple:
+Harbor is the executable contract. ForgeCAD keeps the product code, verifier
+integration, benchmark import tools, and public leaderboard page. The Harbor
+task corpus lives outside this repository.
 
-1. ForgeCAD authors Harbor task folders directly under
-   `ai-labs/reconstruction/tasks/`.
-2. `npm run harbor:reconstruction:refresh` validates those folders and updates
-   `ai-labs/reconstruction/dataset.toml` digests.
-3. Harbor builds the sandbox and runs the agent.
-4. The hidden RewardKit verifier evaluates `/app/submission/main.forge.js`
-   after the agent exits.
-5. RewardKit writes Harbor rewards and per-criterion details; ForgeCAD-specific
-   verifier reports preserve geometry comparison evidence.
+## Repository Split
 
-Harbor is the executable contract for both RL rollouts and benchmark evals.
-There is no second task catalog, no generated sibling registry, and no local
-runner command surface to keep in sync.
-
-The external collaboration repository for SpaceXAI-facing handoff material is:
+Use two neighboring checkouts:
 
 ```text
-https://github.com/KoStard/SpaceXAI_ForgeCAD
+~/Projects/CAD/
+  ForgeCAD/
+    ai-labs/reconstruction-benchmark/
+    src/pages/generatedBenchmarkData.ts
+  ForgeCAD-Harbor-Datasets/
+    reconstruction/
+      dataset.toml
+      agents/
+      tasks/
 ```
 
-Treat that repository as coordination and distribution context. The executable
-benchmark contract is the Harbor dataset in this repo.
+The ForgeCAD repo discovers the external dataset from:
 
-## Current Reconstruction Suite
+```bash
+FORGECAD_RECONSTRUCTION_DATASET_ROOT="$HOME/Projects/CAD/ForgeCAD-Harbor-Datasets/reconstruction"
+```
 
-The canonical reconstruction dataset lives in:
+If the variable is not set, the local tools default to:
 
 ```text
-ai-labs/reconstruction/
+../ForgeCAD-Harbor-Datasets/reconstruction
 ```
 
-The current seed tasks are:
+The external collaboration repository for SpaceXAI-facing handoff material is:
 
 ```text
-ball-bearing
-sealing-frame
-wheel
+https://github.com/KoStard/SpaceXAI_ForgeCAD
 ```
 
-The old source-file-style labels are not primary task IDs. Public benchmark data
-normalizes historical rows to the user-facing labels above.
+Treat that repository as coordination and distribution context. The executable
+task contract is the Harbor dataset.
 
 ## Harbor Task Contract
 
-Each task folder includes:
+Each external task folder includes:
 
 ```text
 README.md
@@ -106,65 +101,33 @@ script. RewardKit writes `/logs/verifier/reward.json` and
 RewardKit output while ensuring `reward.json` has numeric Harbor fields:
 `reward`, `score`, `rawCompareScore`, `guard`, and `valid`.
 
-## NPM Command Surface
+## Command Surface
+
+Run agents through Harbor directly:
+
+```bash
+uvx --from harbor harbor run \
+  -p "$FORGECAD_RECONSTRUCTION_DATASET_ROOT/tasks" \
+  --include-task-name ball-bearing \
+  --agent codex \
+  --model gpt-5.5 \
+  --artifact /app/submission/main.forge.js \
+  --jobs-dir "$HOME/Projects/CAD/ForgeCAD-RL-Agent-Runs/harbor/codex" \
+  --n-concurrent 1
+```
 
-Keep this surface small and boring:
+Use the ForgeCAD repo only for reporting:
 
 ```bash
-npm run harbor:reconstruction:refresh
-npm run harbor:reconstruction:check
-npm run harbor:reconstruction:smoke
-npm run harbor:reconstruction:pack-local
-npm run harbor:reconstruction:clean-local
 npm run benchmark:reconstruction:list
 npm run benchmark:reconstruction:import-harbor -- <job-or-trial-dir> --budget standard
 npm run benchmark:reconstruction:leaderboard
 npm run benchmark:reconstruction:report
+npm run build:benchmark-data
 ```
 
-`refresh` validates task layout and rewrites dataset digests. It does not
-generate task folders. If a task should change, edit the Harbor task folder
-itself.
-
-`check` runs the same validation and fails if `dataset.toml` is stale.
-
-`smoke` runs Harbor with the no-op agent on `ball-bearing`. Use it whenever the
-task image, verifier dependencies, or RewardKit contract changes.
-
-## Local Task Workflow
-
-Use this flow when adding or changing tasks inside the ForgeCAD repository.
-
-1. Add or edit the task under `ai-labs/reconstruction/tasks/<task-id>/`.
-2. Keep the task folder self-contained and Harbor-runnable.
-3. Update the task `README.md` with environment, verifier dimensions, layout,
-   and concrete Harbor run commands.
-4. Refresh the dataset manifest:
-
-   ```bash
-   npm run harbor:reconstruction:refresh
-   npm run harbor:reconstruction:check
-   ```
-
-5. List tasks through the local reporting wrapper:
-
-   ```bash
-   npm run benchmark:reconstruction:list
-   ```
-
-6. Run Harbor directly when testing the lab-facing package:
-
-   ```bash
-   npm run harbor:reconstruction:smoke
-   ```
-
-7. Import completed Harbor jobs when you need local leaderboard/report output:
-
-   ```bash
-   npm run benchmark:reconstruction:import-harbor -- \
-     "$HOME/Projects/CAD/ForgeCAD-RL-Agent-Runs/harbor/smoke/<timestamp>" \
-     --budget smoke
-   ```
+There is no ForgeCAD-owned task sync command. If Harbor can run the external
+dataset path, the benchmark layer can import the resulting jobs.
 
 ## Direct Harbor Runs
 
@@ -176,7 +139,7 @@ built-in Harbor agent:
 ```bash
 CODEX_AUTH_JSON_PATH="$HOME/.codex/auth.json" \
 uvx --from harbor harbor run \
-  -p ai-labs/reconstruction/tasks \
+  -p "$FORGECAD_RECONSTRUCTION_DATASET_ROOT/tasks" \
   --include-task-name ball-bearing \
   --agent codex \
   --model gpt-5.5 \
@@ -188,11 +151,11 @@ uvx --from harbor harbor run \
 Use `--agent-import-path` for a custom adapter such as the bundled Grok adapter:
 
 ```bash
-PYTHONPATH="$PWD/ai-labs/reconstruction/agents" \
+PYTHONPATH="$FORGECAD_RECONSTRUCTION_DATASET_ROOT/agents" \
 GROK_AUTH_JSON="$HOME/.grok/auth.json" \
 GROK_CONFIG_TOML="$HOME/.grok/config.toml" \
 uvx --from harbor harbor run \
-  -p ai-labs/reconstruction/tasks \
+  -p "$FORGECAD_RECONSTRUCTION_DATASET_ROOT/tasks" \
   --include-task-name ball-bearing \
   --agent-import-path grok_harbor_agent:GrokCliAgent \
   --model grok-build \
@@ -201,8 +164,6 @@ uvx --from harbor harbor run \
   --n-concurrent 1
 ```
 
-Current task names are `ball-bearing`, `sealing-frame`, and `wheel`.
-
 ### Full Suite
 
 Codex:
@@ -210,7 +171,7 @@ Codex:
 ```bash
 CODEX_AUTH_JSON_PATH="$HOME/.codex/auth.json" \
 uvx --from harbor harbor run \
-  -p ai-labs/reconstruction/tasks \
+  -p "$FORGECAD_RECONSTRUCTION_DATASET_ROOT/tasks" \
   --agent codex \
   --model gpt-5.5 \
   --jobs-dir "$HOME/Projects/CAD/ForgeCAD-RL-Agent-Runs/harbor/codex" \
@@ -221,11 +182,11 @@ uvx --from harbor harbor run \
 Grok:
 
 ```bash
-PYTHONPATH="$PWD/ai-labs/reconstruction/agents" \
+PYTHONPATH="$FORGECAD_RECONSTRUCTION_DATASET_ROOT/agents" \
 GROK_AUTH_JSON="$HOME/.grok/auth.json" \
 GROK_CONFIG_TOML="$HOME/.grok/config.toml" \
 uvx --from harbor harbor run \
-  -p ai-labs/reconstruction/tasks \
+  -p "$FORGECAD_RECONSTRUCTION_DATASET_ROOT/tasks" \
   --agent-import-path grok_harbor_agent:GrokCliAgent \
   --model grok-build \
   --jobs-dir "$HOME/Projects/CAD/ForgeCAD-RL-Agent-Runs/harbor/grok" \
@@ -237,150 +198,25 @@ Direct Harbor CLI-agent runs need outbound internet from the task container for
 agent install/auth/model calls, so the task definitions set
 `allow_internet = true`. The verifier remains hidden until the agent exits.
 
-## Remote Harbor Runs On popos
-
-Use `popos` as a remote Harbor execution host when local machine resources are
-the bottleneck. Do not add a second benchmark runner for this. The remote
-machine should run the same Harbor commands against the same task folders.
-
-Remote prerequisites:
-
-1. `ssh popos` reaches the server.
-2. Docker, `uvx`, Node, and npm are installed on the server.
-3. A dedicated checkout exists at `~/Projects/CAD/ForgeCAD`.
-4. Agent auth lives outside the repo on the server.
-5. Run outputs stay outside the checkout under
-   `~/Projects/CAD/ForgeCAD-RL-Agent-Runs/harbor/`.
-
-For normal runs, update the dedicated server checkout with git after pushing
-the branch you want to test:
-
-```bash
-branch="$(git branch --show-current)"
-git push origin "$branch"
-ssh popos "cd ~/Projects/CAD/ForgeCAD && \
-  git fetch origin && \
-  git switch '$branch' && \
-  git pull --ff-only origin '$branch'"
-```
-
-Use `rsync` only for scratch experiments where the local work is intentionally
-uncommitted. It mirrors the dirty checkout onto the server, so use it only
-against a dedicated throwaway checkout:
-
-```bash
-rsync -az --delete \
-  --exclude .git \
-  --exclude node_modules \
-  --exclude dist \
-  --exclude dist-cli \
-  --exclude dist-bundles \
-  --exclude .data \
-  --exclude 'ai-labs/reconstruction/tasks/*/environment/forgecad-package/*.tgz' \
-  ./ popos:~/Projects/CAD/ForgeCAD/
-```
-
-Run preflight validation on the server:
-
-```bash
-ssh popos 'cd ~/Projects/CAD/ForgeCAD && npm run harbor:reconstruction:check'
-```
-
-If `popos` reports `Missing script: "harbor:reconstruction:check"`, the server
-checkout is stale. Update the server checkout with git first.
-
-Run the no-op Harbor smoke on the server:
-
-```bash
-ssh popos 'cd ~/Projects/CAD/ForgeCAD && npm run harbor:reconstruction:smoke'
-```
-
-Run one task on one Codex agent remotely:
-
-```bash
-ssh popos 'cd ~/Projects/CAD/ForgeCAD && \
-CODEX_AUTH_JSON_PATH="$HOME/.codex/auth.json" \
-uvx --from harbor harbor run \
-  -p ai-labs/reconstruction/tasks \
-  --include-task-name ball-bearing \
-  --agent codex \
-  --model gpt-5.5 \
-  --artifact /app/submission/main.forge.js \
-  --jobs-dir "$HOME/Projects/CAD/ForgeCAD-RL-Agent-Runs/harbor/codex" \
-  --n-concurrent 1'
-```
+## Run Data And Leaderboard
 
-Run one task on the bundled Grok adapter remotely:
-
-```bash
-ssh popos 'cd ~/Projects/CAD/ForgeCAD && \
-PYTHONPATH="$PWD/ai-labs/reconstruction/agents" \
-GROK_AUTH_JSON="$HOME/.grok/auth.json" \
-GROK_CONFIG_TOML="$HOME/.grok/config.toml" \
-uvx --from harbor harbor run \
-  -p ai-labs/reconstruction/tasks \
-  --include-task-name ball-bearing \
-  --agent-import-path grok_harbor_agent:GrokCliAgent \
-  --model grok-build \
-  --artifact /app/submission/main.forge.js \
-  --jobs-dir "$HOME/Projects/CAD/ForgeCAD-RL-Agent-Runs/harbor/grok" \
-  --n-concurrent 1'
-```
-
-Fetch remote Harbor jobs back before importing them into the local benchmark
-ledger:
-
-```bash
-rsync -az \
-  popos:~/Projects/CAD/ForgeCAD-RL-Agent-Runs/harbor/ \
-  "$HOME/Projects/CAD/ForgeCAD-RL-Agent-Runs/harbor/"
-
-npm run benchmark:reconstruction:import-harbor -- \
-  "$HOME/Projects/CAD/ForgeCAD-RL-Agent-Runs/harbor/codex/<timestamp>" \
-  --budget standard
-```
-
-By default, task Dockerfiles install the pinned published ForgeCAD npm package.
-For unpublished local ForgeCAD changes, run the local package flow on the server
-checkout:
-
-```bash
-ssh popos 'cd ~/Projects/CAD/ForgeCAD && npm run harbor:reconstruction:pack-local'
-ssh popos 'cd ~/Projects/CAD/ForgeCAD && npm run harbor:reconstruction:smoke'
-ssh popos 'cd ~/Projects/CAD/ForgeCAD && npm run harbor:reconstruction:clean-local'
-```
-
-If `pack-local` fails because repo build tools are missing, run `npm install`
-once in the server checkout.
-
-There is intentionally no `popos` npm script. The host alias, checkout path,
-and auth file locations are per-developer machine configuration; the repo-owned
-commands remain the small Harbor command surface documented above.
-
-## Local Reporting Wrapper
-
-The local reporting wrapper is not a runner. It exists only to list Harbor
-tasks, import completed Harbor jobs, and build submission ledgers, leaderboards,
-and visual reports.
-
-There is no local matrix command, no host-side agent adapter layer, no separate
-task tracker, and no host-side evaluator. If an agent should run, run it
-through Harbor.
-
-## Run Data
-
-Keep run outputs outside the source checkout under:
+Keep run outputs outside the source checkout:
 
 ```text
 ~/Projects/CAD/ForgeCAD-RL-Agent-Runs/
-  benchmark/
   harbor/
-  research/
-  smoke/
+    <agent>/<job>/
+  benchmark/main/
+    submissions.jsonl
+    submissions/
+    leaderboard.json
+    leaderboard.md
+    report/
 ```
 
-Keep raw agent logs and Harbor job directories. They are evidence for prompt
-design, task packaging, tool behavior, and reward design.
+Raw Harbor jobs are evidence. Keep them somewhere durable, but do not commit
+them to ForgeCAD. They contain logs, generated code, renders, verifier payloads,
+local paths, and provider-specific artifacts.
 
 Important Harbor files:
 
@@ -397,6 +233,22 @@ Important Harbor files:
   verifier/reference.png
 ```
 
+The public leaderboard is updated by importing selected Harbor jobs into the
+local benchmark ledger, regenerating reports, then regenerating the repo-safe
+snapshot:
+
+```bash
+npm run benchmark:reconstruction:import-harbor -- \
+  "$HOME/Projects/CAD/ForgeCAD-RL-Agent-Runs/harbor/codex/<job>" \
+  --budget standard
+npm run benchmark:reconstruction:report
+npm run build:benchmark-data
+```
+
+Commit `src/pages/generatedBenchmarkData.ts` to publish the website update.
+The committed snapshot contains only sanitized task, model, score, reward,
+metric, and timing fields. It does not contain raw logs or private paths.
+
 ## Result Payload Contract
 
 The Harbor verifier writes numeric reward files:
@@ -430,32 +282,25 @@ For reconstruction tasks, the geometry component should include:
 - Hard caps for obvious shortcut failures.
 
 Avoid solving score problems by only changing a curve shape. If a two-ring
-bearing approximation scores too high, the missing signal is structural: rolling
-elements, feature edges, component count, section occupancy, or topology. Add
-the missing measurement and then calibrate the scalar projection.
+bearing approximation scores too high, the missing signal is structural:
+rolling elements, feature edges, component count, section occupancy, or
+topology. Add the missing measurement and then calibrate the scalar projection.
 
 ## Maintaining Tasks
 
-When adding or changing a task:
-
-1. Put the task under `ai-labs/reconstruction/tasks/<task-id>/`.
-2. Use a user-facing slug such as `ball-bearing`, not an opaque source asset ID.
-3. Include the full Harbor layout: prompt, task TOML, environment, reference,
-   starter, tests, task metadata, and README.
-4. Record the reference asset SHA-256 in `tests/task.json`.
-5. Keep `/app/submission/main.forge.js` as the stable deliverable.
-6. Refresh the dataset with `npm run harbor:reconstruction:refresh`.
-7. Run identity scoring: reference vs reference should score near 100.
-8. Run the starter score and record it.
-9. Run at least one known weak shortcut candidate and one stronger candidate.
-10. Confirm the stronger candidate ranks higher for the right reasons.
-11. Store local experiment artifacts under `docs/temporary/projects/...`, then
-    promote only durable conclusions here.
+When adding or changing a task, edit the external Harbor dataset. Keep task IDs
+user-facing, include the full Harbor layout, record reference SHA-256 in
+`tests/task.json`, and keep `/app/submission/main.forge.js` as the stable
+deliverable.
+
+Run enough Harbor jobs to prove the task behaves correctly: reference identity
+scoring, starter scoring, at least one weak shortcut candidate, and one stronger
+candidate. The stronger candidate should rank higher for the right reasons.
 
 ## Maintaining Agent Adapters
 
-Agent adapters belong in Harbor, not in the local benchmark reporting wrapper.
-For each supported Harbor agent CLI, keep a short note answering:
+Agent adapters belong with the external Harbor dataset. For each supported
+Harbor agent CLI, keep a short note answering:
 
 1. How to run headlessly.
 2. How to set cwd.
@@ -492,17 +337,14 @@ saved runs.
 Before merging reconstruction environment changes:
 
 1. `npm test` passes, or a narrower check is justified in the PR.
-2. `npm run harbor:reconstruction:refresh` passes.
-3. `npm run harbor:reconstruction:check` passes.
-4. `npm run benchmark:reconstruction:list` passes.
-5. `npm run harbor:reconstruction:smoke` passes when environment packaging
-   changes.
-6. Verifier reward output contains numeric `reward`, `score`,
+2. `npm run benchmark:reconstruction:list` passes against the external dataset.
+3. Harbor can run the changed task directly from the external dataset.
+4. Verifier reward output contains numeric `reward`, `score`,
    `rawCompareScore`, `guard`, and `valid`.
-7. Verifier report output contains the detailed status, score vector,
+5. Verifier report output contains the detailed status, score vector,
    aggregation, guard data, and artifacts.
-8. RewardKit writes `reward-details.json`.
-9. Agent logs are captured in the Harbor trial folder.
-10. Timeouts are enforced by Harbor and inside the verifier.
-11. No global skills or parent repo instructions are required.
-12. The docs index and this guide are updated when the contract changes.
+6. RewardKit writes `reward-details.json`.
+7. Agent logs are captured in the Harbor trial folder.
+8. Timeouts are enforced by Harbor and inside the verifier.
+9. No global skills or parent repo instructions are required.
+10. The docs index and this guide are updated when the contract changes.

package/dist/docs-raw/skills/forgecad-3d-reconstruction.md

@@ -36,13 +36,13 @@ Do not solve reconstruction by returning `importMesh("reference.stl")` or `impor
    No wrapper script is needed. Use the local checkout CLI:
 
    ```bash
-   node dist-cli/forgecad.js run path/to/source.stl --quality live --details
+   node dist-cli/forgecad.js ls path/to/source.stl --quality live --long
    node dist-cli/forgecad.js render 3d path/to/source.stl /tmp/<slug>-source.png --camera iso --edges thin --size 900
    node dist-cli/forgecad.js inspect visual objects path/to/source.stl /tmp/<slug>-source-objects --camera iso --size 700 --force
    node dist-cli/forgecad.js inspect sections sample path/to/source.stl /tmp/<slug>-source-sections --count 5 --size 700 --force
    ```
 
-   For 3MF sources, `run --details` also prints the source archive's build
+   For 3MF sources, `forgecad run` prints the source archive's build
    items/resource objects with stable refs such as
    `3mf:build:001:object:2`, automatic names, per-item bounding boxes, and
    triangle counts. Use that item table to avoid missing hidden multi-part

package/dist/docs-raw/skills/forgecad-make-a-model.md

@@ -65,7 +65,7 @@ Use today's date for the directory. Use the user's current ForgeCAD project when
    - Return the final geometry (single shape, array, or named objects array)
    - Treat `fillet(shape, r)` and `chamfer(shape, r)` as experimental edge treatments: Manifold can produce incorrect results and OCCT can be very slow. Prefer simpler primitive profiles, lower segment counts, targeted edge selectors, and inspection before relying on the result.
 4. Validate — run `forgecad run <file>` to check for errors. For multi-file projects, always validate `main.forge.js`.
-5. Verify geometry — render a multi-angle visual evidence set before final delivery: whole-model context plus agent-chosen orthographic, oblique, underside, or hidden-object views that expose the relevant components and interfaces. Choose camera directions from the model's shape and likely failure modes, not from a fixed recipe. Use those views to look for internals that are accidentally visible, parts that visibly do not fit, floating details, blocked access, missing seats, and unexpected interference. Run `forgecad run --connectivity` when the model has multiple returned objects or visible attachments, run `forgecad debug assembly --fail-on warning` when the script uses `assembly()`, run `forgecad inspect mechanical-integrity <project-or-file> --collisions` before sharing generated mechanical work, and run the targeted `forgecad inspect <family> <mode>` commands that match the task (see Final Acceptance Gate and Render-Verify Loop below). For multi-file projects, render and inspect `main.forge.js`. Collision findings are model work, not FYI: remove unexpected overlaps before delivery.
+5. Verify geometry — render a multi-angle visual evidence set before final delivery: whole-model context plus agent-chosen orthographic, oblique, underside, or hidden-object views that expose the relevant components and interfaces. Choose camera directions from the model's shape and likely failure modes, not from a fixed recipe. Use those views to look for internals that are accidentally visible, parts that visibly do not fit, floating details, blocked access, missing seats, and unexpected interference. Run `forgecad inspect physical components` when the model has multiple returned objects or visible attachments, run `forgecad debug assembly --fail-on warning` when the script uses `assembly()`, run `forgecad inspect mechanical-integrity <project-or-file> --collisions` before sharing generated mechanical work, and run the targeted `forgecad inspect <family> <mode>` commands that match the task (see Final Acceptance Gate and Render-Verify Loop below). For multi-file projects, render and inspect `main.forge.js`. Collision findings are model work, not FYI: remove unexpected overlaps before delivery.
 6. Iterate from visual and inspection feedback — treat every render and inspection bundle as model evidence, not a checkbox. Read the normal PNGs, manifest, and evidence PNGs; convert each unexpected collision, thin region, missing section detail, wrong component count, floating body, distance gap, confusing object-color result, accidentally exposed internal structure, bad fit, or visually unsupported interface into a concrete model edit; then rerun the same targeted evidence pass until the result matches the intended physical component graph.
 
 ### Manufacturing Process Is Not Assumed
@@ -188,7 +188,7 @@ Before telling the user the model is done, prove both technical validity and vis
 1. State the intended physical component graph. Decide whether the final artifact should be one connected component, several intentionally separate components, or a selected assembly plus named ghosts. Then run:
 
    ```bash
-   forgecad run model.forge.js --connectivity
+   forgecad inspect physical components model.forge.js --camera iso
    ```
 
    The reported component count must match the design intent. Treat unexpected islands, accidental fusion, or bbox-only "touching" that does not make physical sense as model bugs.
@@ -297,7 +297,7 @@ For important components, collect both:
 - Context view — neighbors present, proving the part belongs in the final assembly.
 - Focus view — only the relevant objects visible, making small gaps, intersections, missing seats, and floating parts easy to see.
 
-Prefer CLI `--focus` / `--hide` filters, named views, or parameter-selected diagnostic modes over changing production geometry. Use the object names from `node dist-cli/forgecad.js run model.forge.js --quality live` when you are unsure what the filters should target.
+Prefer CLI `--focus` / `--hide` filters, named views, or parameter-selected diagnostic modes over changing production geometry. Use the object paths from `node dist-cli/forgecad.js ls model.forge.js --tree` when you are unsure what the filters should target.
 
 #### Structured inspection bundles
 

package/dist/docs-raw/skills/forgecad-reconstruction-benchmark.md

@@ -77,13 +77,13 @@ ForgeCAD commands.
    inference.
 
    ```bash
-   ./bin/forgecad run task/reference/<asset> --quality live --details
+   ./bin/forgecad ls task/reference/<asset> --quality live --long
    ./bin/forgecad render 3d task/reference/<asset> outputs/reference.png --camera iso --edges thin --size 900
    ./bin/forgecad inspect sections sample task/reference/<asset> outputs/reference-sections --count 5 --size 700
    ./bin/forgecad inspect section task/reference/<asset> --plane yz --offset 0 --ray width:-50,0:50,0 --size 700
    ```
 
-   For 3MF references, the `run --details` output includes a source structure
+   For 3MF references, the `forgecad run` output includes a source structure
    table with stable `3mf:build:...:object:...` refs, automatic item names,
    per-item bounding boxes, and triangle counts. Treat that table as part of
    the evidence: the final model may be one part or many parts, but you should
@@ -93,7 +93,7 @@ ForgeCAD commands.
    unique probe directory with `result.json`; later you can replay that probe
    against the candidate with `./bin/forgecad inspect replay <result.json>
    --source submission/main.forge.js`.
-   For 3MF references, the `run --details` output includes a source structure
+   For 3MF references, the `forgecad run` output includes a source structure
    table with stable `3mf:build:...:object:...` refs, automatic item names,
    per-item bounding boxes, and triangle counts. Treat that table as part of
    the evidence: the final model may be one part or many parts, but you should

package/dist/index.html

@@ -55,7 +55,7 @@
       * { margin: 0; padding: 0; box-sizing: border-box; }
       html, body, #root { width: 100%; min-height: 100%; background: var(--fc-bg); color: var(--fc-text); font-family: system-ui, -apple-system, sans-serif; }
     </style>
-    <script type="module" crossorigin src="/assets/app-BUZqJvSO.js"></script>
+    <script type="module" crossorigin src="/assets/app-bEww1ic4.js"></script>
     <link rel="stylesheet" crossorigin href="/assets/app-CsHnaBWt.css">
   </head>
   <body>

package/dist/sitemap.xml

@@ -2,43 +2,43 @@
 <urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
   <url>
     <loc>https://forgecad.io/</loc>
-    <lastmod>2026-05-29</lastmod>
+    <lastmod>2026-05-31</lastmod>
     <changefreq>weekly</changefreq>
     <priority>1.0</priority>
   </url>
   <url>
     <loc>https://forgecad.io/docs</loc>
-    <lastmod>2026-05-29</lastmod>
+    <lastmod>2026-05-31</lastmod>
     <changefreq>weekly</changefreq>
     <priority>0.8</priority>
   </url>
   <url>
     <loc>https://forgecad.io/benchmark</loc>
-    <lastmod>2026-05-29</lastmod>
+    <lastmod>2026-05-31</lastmod>
     <changefreq>weekly</changefreq>
     <priority>0.8</priority>
   </url>
   <url>
     <loc>https://forgecad.io/blog</loc>
-    <lastmod>2026-05-29</lastmod>
+    <lastmod>2026-05-31</lastmod>
     <changefreq>weekly</changefreq>
     <priority>0.7</priority>
   </url>
   <url>
     <loc>https://forgecad.io/pricing</loc>
-    <lastmod>2026-05-29</lastmod>
+    <lastmod>2026-05-31</lastmod>
     <changefreq>monthly</changefreq>
     <priority>0.6</priority>
   </url>
   <url>
     <loc>https://forgecad.io/examples</loc>
-    <lastmod>2026-05-29</lastmod>
+    <lastmod>2026-05-31</lastmod>
     <changefreq>weekly</changefreq>
     <priority>0.6</priority>
   </url>
   <url>
     <loc>https://forgecad.io/blog/hello-forgecad-io</loc>
-    <lastmod>2026-05-29</lastmod>
+    <lastmod>2026-05-31</lastmod>
     <changefreq>monthly</changefreq>
     <priority>0.5</priority>
   </url>

package/dist-cli/{check-compiler-LOXCPEOI.js → check-compiler-U5SOPN7X.js}

@@ -9,7 +9,7 @@ import {
   resolvePackagePath,
   runDirectCliMain,
   setActiveBackend
-} from "./chunk-RY43WF46.js";
+} from "./chunk-EXWGNL6K.js";
 
 // cli/check-compiler.ts
 import assert from "assert/strict";
@@ -636,4 +636,4 @@ runDirectCliMain(import.meta.url, "cli/check-compiler.ts", () => runCheckCompile
 export {
   runCheckCompilerCli
 };
-//# sourceMappingURL=check-compiler-LOXCPEOI.js.map
+//# sourceMappingURL=check-compiler-U5SOPN7X.js.map

package/dist-cli/{check-query-propagation-BAKNVWXR.js → check-query-propagation-XOKNSSYU.js}

@@ -12,7 +12,7 @@ import {
   resolvePackagePath,
   runDirectCliMain,
   setActiveBackend
-} from "./chunk-RY43WF46.js";
+} from "./chunk-EXWGNL6K.js";
 
 // cli/check-query-propagation.ts
 import assert from "assert/strict";
@@ -488,4 +488,4 @@ runDirectCliMain(import.meta.url, "cli/check-query-propagation.ts", () => runChe
 export {
   runCheckQueryPropagationCli
 };
-//# sourceMappingURL=check-query-propagation-BAKNVWXR.js.map
+//# sourceMappingURL=check-query-propagation-XOKNSSYU.js.map