* claim: #4083
* Feature editors refill named edges and faces where they are now (#4083)
After an upstream sketch change, Fillet, Chamfer, Shell and Draft evaluated
correctly, but their editors refilled the originally clicked points and
could then fail with "no edge at ...". Resolve the stored names through the
document's naming helpers first, keep the resolved index, and map the pick
into the displayed body's frame; the point lookup stays as the fallback.
* claim: #4097
* Sketch fillet keeps a corner placed by an expression (#4097)
Filleting a corner whose position comes from parameter expressions deleted
the point and its expressions, so the sketch gained two degrees of freedom
and later parameter changes distorted the outline. keep_corner now also
keeps fixed or expression-positioned points as virtual sharp corners.
* claim: #4150
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Browser: a second click on a Document Settings entry closes its window (#4150)
Each entry under Document Settings only opened its window. A second click on
Units closed the menu on the press (the row is outside it) and opened it again
on the release; Precision left Preferences open; Parameters and Design
Properties came back as a fresh copy. Now a click on an entry whose window is
open closes it the way the window's own x does:
- Units: context_menu::toggle_for with the row as the opener, the fix#1522
made for the workspace button's menu. A right-click still opens it at the
pointer.
- Precision: prefs::close(app, true), the Cancel/x path that puts back what
changed since the window opened (now pub(crate)).
- Parameters and Design Properties: the window is dropped, as its x does. A
Parameters value typed but not applied yet is dropped with it, as today's
second click already drops it.
Fixes#4150
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4353
* Renaming a plane keeps its projections, canvases and hidden state (#4353)
timeline.rename only renamed the feature in other features' references. A sketch's
projection of a construction plane, a canvas placed on it and the hidden list keep it
by name too, so they lost it: the sketch warned that the plane no longer exists, the
canvas left the view and a hidden plane showed again. Document::rename_feature_keys
renames it in sketch links and canvases, as rename_body_keys does for bodies; the
selection follows in the command, and the hidden list in Session::refresh_from, so a
hidden plane stays hidden through undo and redo too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5269
* DXF: write and read a Bézier at its own degree (#5269)
The DXF writer and reader held a spline's degree to 7, the value MAX_BEZIER_DEGREE had before
#1642 raised it to 9. A G4 blend (one Bézier of degree 9) was written as a degree-7 SPLINE over
its first 8 control points, so a flat pattern's cut-out stopped 4.444 mm short in the file.
- seg_shape takes its limit from MAX_BEZIER_DEGREE: a Bézier is written whole, at its own degree.
- read_dxf reads SPLINE degrees 1 to MAX_BEZIER_DEGREE, the degrees a sketch spline takes.
- Tests: every Bézier degree 4 to 9, a degree-9 round trip, the reader's degrees, hostile SPLINE
degrees and too few control points, the G4 cut-out exported and read back with its areas, and a
degree-9 SPLINE with 2 to 10 points inserted into a sketch.
Fixes#5269
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3989
* ui-egui tests: offscreen GPU tests start their devices one at a time
Every offscreen test started a wgpu instance of its own, and starting one
makes the Vulkan loader scan its drivers. With several test threads two
tests doing that at once crashed the test process with SIGSEGV inside the
loader (loader_icd_scan). offscreen_device now holds a process-wide lock
while it starts the instance, adapter and device; each test keeps its own
device and the rendering still runs side by side.
A new Graphics preference, "Show reference geometry through bodies", draws
inactive sketch contours, origin and construction planes and their axes over
shaded bodies. It is off by default, so the viewport looks as before: reference
geometry is depth-tested and a section leaves it whole (#1229); Create Sketch
and dialogs picking a plane still draw the planes through the model.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#5109 added Frame::hover_tint and selection_tint; two test frames that landed alongside it didn't set them, so solidcraft-gpu's tests stopped compiling on main.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch on a picked origin plane uses the plane shown, also in a moved component
After a STEP file opens its component stays the active one, and standing it up
turns that component a quarter about X. Starting a sketch on XY picked in the
viewport then sent `{"plane": "XY"}`, which names the active component's own
XY: upright after the stand-up. The sketch landed on the world XZ, Look At
turned to face Y and the extrude went along Y.
The viewport and browser only show the design's origin planes, so a picked
origin plane now goes to sketch.create written out as that world plane while
the active component is moved; sketch.create maps it into the component's
frame once. The flip toward the viewer uses the same plane.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Stand-up axes tests: stand the file up through the offer when one is waiting
#5655 offers to stand up SolidWorks files instead of turning them silently.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Hovering or selecting whole bodies -- a browser row, a History group, a body
under the pointer in body-pick mode -- built a CPU copy of every body's
triangles into the highlight scene. On a 40-body 4.3 MB STEP assembly that was
147 bodies, 70 MB of triangle copies and a 519 ms frame.
The model already has the same geometry on the GPU. The highlight now names it
and the renderer draws it again in the highlight colour:
- bodies drawn from a display plan: the model's evaluated plan chunks, with a
hover/selection tint uniform (the chunks stay resident in the model scene);
- bodies drawn as placements of shared triangles: the same placements of the
model's shared triangles, coloured per instance;
- bodies the model builds on the CPU: their vertex range in the model scene's
block.
See-through and faded bodies keep the CPU feedback, which colours them; faces
and edges highlighted on their own are unchanged. The highlight looks the same
(verified side by side).
On the assembly the highlight build drops from 519 ms to 9 ms and no longer
copies any triangles; selecting the whole assembly takes 9-20 ms.
Co-authored-by: Claude Code <noreply@anthropic.com>
* claim: #5295
* An empty intersect is refused instead of deleting the body (#5295)
An intersect whose boolean came out empty removed the target body and
reported success. The shared evaluator now refuses it when nothing of a
named target would remain, when nothing of any target would remain, or
when the tools meet a target and the result is still empty. An unnamed
target the tools miss altogether is still removed, as an intersect with
every body does (#5160). Sheet metal flat machining follows the same rule.
* Sheet metal: an intersect with already removed material is refused (#5295)
* claim: #133
* mesh: Save As Mesh declares the panel that shows it (#133)
The ribbon redesign (#2176) removed the MESH tab, but three things still
pointed at it:
- `file.save_mesh` declared `.at("MESH", "MODIFY")`. The command inventory
prints the placement a spec declares, so `docs/features.md` advertised a
`MESH > MODIFY` location no ribbon has. Its real home is already
INSERT / FILES (`SECTIONS` lists it), so the declaration now agrees with
the ribbon; `destination()` resolved it there before and after, so nothing
moves on screen.
- The user guide sent people to "the Mesh tab under Modify" for Save as mesh.
- A test now refuses any command that declares a tab the ribbon removed, and
pins where Save As Mesh lives.
`PLACEMENTS` keeps its MESH entries: those migrate toolbars users saved under
the old panels, which is a different question from what a spec declares.
---------
Co-authored-by: Geraldrunion <geraldrunion@users.noreply.github.com>
* claim: #4563
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Section Analysis: a Hatch option, off for a plain cut in the body's colour (#4563)
The Section Analysis dialog gets a Hatch check box beside Flip, and
inspect.section a hatch?: bool (default true), so scripts, the CLI and the
control channel reach it too. Hatched, a cut is its body's colour tinted
toward the cap colour with 45 degree stripes, as before; off, it is the
body's own solid colour. The viewport (cap alpha in the uniform), the CPU
rasterizer and ui.render follow it, and the section keeps it when hidden
and shown again or edited.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4589
* Expression suggestions stay closed while another control is dragged (#4589)
A drag elsewhere leaves a text box its focus (egui gives it up on a click,
not a drag), so dragging the Drive Joints slider rewrote the focused Angle
box under its old selection. A letter before the caret ("5.0 d|eg") made a
new word, a fresh undismissed popup, and the list opened over the slider.
Dismissed suggestions now stay closed until the box is typed in, a drag of
another widget dismisses them, and a hidden list forgets its row rects.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4609
* Mesh Convert works without activating the component first (#4609)
Inserting a mesh puts it in a new component while the parent stays
active, so Convert Mesh always refused the first time. Convert now
resolves the mesh body's owning component and adds the conversion
there when the active component is that component or an ancestor,
leaving the active component unchanged. From another branch it still
refuses, and the error names the component to activate.
* Control: a whole-design measure reads a copy off the window thread
`inspect.measure` and `document.inspect` (which measures every body by
default) ran on the window thread. On a large assembly that took about a
minute, the window answered nothing meanwhile, and Windows closed it as
not responding (#4604). They change nothing, so they now run on a copy of
the session on a worker and are answered from there, like exports, but
whatever the "agent commands off the window thread" preference says: an
answer from the design as it was when asked has no order to keep. The
window keeps answering `ui.*` and drawing meanwhile.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Control: clearance and interference checks read a copy off the window thread too
inspect.clearance and inspect.interference only read the design, and on a
large assembly measure every pair of bodies: on a 153-body design a clearance
check computed for over five minutes on the window thread, the window showing
"not responding". They join inspect.measure and document.inspect in running on
a copy, answered from there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4030
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch Symmetry: a dimensioned shape slides onto the axis (#4030)
With a rectangle's width dimensioned, Symmetry of its two sides about an
axis was refused: the second side can't move on its own, the axis never
moves, and the ordinary solve that slides the whole rectangle only ran
when every first-picked entity could move. One movable first entity is
now enough.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4328
* Fix sheet conversion picks on tangent bend edges (#4328)
Resolve cylindrical bends through the nearest edge's incident faces when a
pick on a tangent boundary resolves to its planar panel. This keeps valid
remote bends connected, including picks on the other skin of a shell.
Keep conversion geometry and picks in the body's local frame, and format
conversion diagnostics through its world placement. Cover rotated bend
picks and disconnected-pick coordinates with synthetic regressions.
A join failed with the kernel's internal error "tolerance must be no less
than 1e-6" at vendor/truck-geometry/src/decorators/intersection_curve.rs:172:
IntersectionCurve::parameter_division asserted truck's minimum tolerance,
and the guard turned the panic into a failed operation. A tolerance below
truck's TOLERANCE, or one that is not a number, now divides at TOLERANCE:
the curve's points are only found to within it anyway.
The test divides the intersection edges of a hole cut through a B-spline
face at 5e-7, 1e-12, 0, -1 and NaN: each gives the division at 1e-6. Without
the fix it panics with the reported message at the reported line.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4593
* Pattern and mirror: a rib as the source (#4593)
A pattern or mirror of features refused a rib as a source ("Rib1
cannot be patterned or mirrored yet"). A rib's walls run until they
meet the body, so each copy is built again where it is: the rib's
sketch is moved by the copy's transform and the walls are run against
the bodies there, as a hole is drilled again at each copy. The copies
join the body, as the rib does. The rib's own tool building moves into
rib_tools, which the rib feature calls unmoved.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch Point: a point put on a midpoint or a curve snap is held there (#4691)
The Point tool sent only a snapped point's coordinates and sketch.point
took only [x, y], so a point clicked on a line's midpoint (or on a curve,
or a model vertex) was free to move off it. sketch.point now takes the
snap forms the other draw commands take ("mid:<line>", "on:<curve>:x,y",
{vertex}, {on_edge}); the snap makes the point with its constraint. The
tool passes those snaps as themselves; a snap to an existing point still
goes as coordinates, where `connect` joins them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch Point: refuse a snap it can't read instead of answering as if it made a point
sketch_project::snap_points leaves a snap it can't parse ("on:c1:abc",
"on:c1", "on:c1:1,2,3") as it was given, and sketch.point then read
that string back as the point it had made: no error, no point, and an id
a script would go on to dimension. It now takes the snap only when it
names a point of the sketch, and otherwise refuses, listing the snap
forms (found by the test agent on #4745).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The app's background loader (file_load::control) kept only `up` and
`names` of a control-channel open or import, so file.insert_step's
`matrix` and `name` were dropped and the part landed at the origin, and
file.insert_mesh lost all its options. It keeps every parameter but the
path, as given; each command checks them as it does when run directly.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
timeline.rollback / timeline.roll_to read a negative position as 0, so
an agent passing -1 for "one step back" rolled the whole timeline back
and every body disappeared, with no error. A negative position is now
refused. The two commands shared one function that refused in
roll_to's name; each now refuses in its own.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Nine messages: the pattern tools, their headings and two sample
descriptions. "Réseau" can be mistaken for a network.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #4452
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Combine takes bodies clicked in the Browser (#4452)
A body row in the Browser only replaced the selection, so with Combine
open its target and tools stayed empty and overlapping bodies could only
be picked in the view. When the open dialog's active input takes bodies,
a body row now goes there, as plane and axis rows already do.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Combine Browser picks: test both layouts and control-channel clicks (#4452)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The Keyboard Shortcuts page refused a key another command used, even a
sketch tool and a model command that never run in the same mode, so
Shift+S could not open Create Sketch in the model and place a point in a
sketch. Edited by hand, the press ran Create Sketch inside the sketch.
A sketch tool (SKETCH tab) and a model command (the other tabs) may now
share a key. A press runs the enabled command of the mode the view is in
first: the tool in a sketch, the model command outside. Two commands of
one mode still clash, and commands on no tab (undo, views, files) clash
with everything.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4783
* Physical material: one rule for a body's material and its density (#4783)
inspect.measure, component.properties, occurrence.properties and the
Properties window took a body's material name from the raw assignment,
but its density from Document::density, which falls back to the default
material when the assigned one is unknown or has an unusable density. So
a damaged or hand-edited design read one material and weighed another: a
design material "Broken" at -3 g/cm³ read "Broken" at 1.29 g/cm³.
One rule, in one place:
- DEFAULT_MATERIAL (new, beside DEFAULT_DENSITY in the material library)
is the default material's name. The library's first row uses it.
- Document::body_material (new, beside Document::density) is the material
a body's mass is computed with, as the library or the design spells it:
the assigned one when it is known and has a usable density, else
DEFAULT_MATERIAL. Document::density is that material's density, with
the same value as before for every input.
inspect.measure and the physical properties name body_material.
inspect.measure now names every body's material, measured or not, since
naming it takes no measuring; the density stays with the mass, on the
bodies measured. So the Properties window shows the engine's name without
a fallback of its own. The places that keep the assignment as stored
(configuration cells, material.assign, material.edit, material.delete
and the dialogs) only use the constant, so a configuration still keeps a
damaged assignment as it is.
The visible change: a body whose material is unknown, or has an unusable
density, reads "Default", the material it weighs, and a miscased name
reads as the library or the design spells it. A body inspect.measure
can't measure now has its material too. Masses and densities don't
change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The Move dialog's floating value box always showed and edited the X
distance, whichever arrow was dragged, and Tab left the box for another
widget. Now:
- dragging an arrow shows that axis's distance, turning a ring (Free Move)
that axis's angle; a plane square or the ball leaves the box as it is;
- Tab in the box goes to the next value (X, Y, Z, then in Free Move the
X, Y and Z angles) and Shift+Tab back, the keyboard staying in the box.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2987
* Flange on a plain solid says how to start sheet metal (#2987)
Flange, Flat Pattern, and Export DXF on a design with no sheet metal body
now say how to make one: Flange > Base, or Convert to Sheet Metal on a
plate's large flat face. An edge flange checks for a sheet metal body
before it adds a feature. The message is in the catalogs, and engine
messages that match a catalog entry are shown in the interface language.
* Dialogs show a refusal without the engine's parameter prefix (#2987)
The Flange dialog showed "invalid parameters for `sheet.flange`: " in front
of the new guidance. The dialog already names its command, so the red
refusal text leaves that prefix out and follows the interface language
when the message is in the catalogs. Create Issue still gets the full text.
* claim: #4116
* Web: a trackpad swipe pans instead of zooming (#4116)
Browsers deliver trackpad swipes as wheel events without scroll phases, so
the shared navigation treated every vertical swipe as a mouse-wheel zoom.
The web raw-input hook now brackets point-mode swipes (small or horizontal
deltas) with gesture start/end phases, ending after 100 ms idle or before
line/page wheel or pinch input. Wheel notches and pinch still zoom, and
camera input still goes through navigation::handle.
* claim: #221
* Units of new designs apply to the blank design already open (#221)
Changing Preferences > Units of new designs only took effect at the next
File > New, so the empty design open at launch stayed in millimetres. When
the preference changes (the dialog, Cancel, Restore Defaults, or
preferences.set) and the open design is blank and unsaved, it takes the new
units at once, without an undo step or unsaved changes. A design with
content or a file keeps its units.
* claim: #4794
* Hem: the direction arrow points the way the hem builds (#4794)
The Hem dialog shared the edge flange's arrow, which points out of the
sheet's larger neighbouring face. A hem folds back the other way, so the
arrow pointed away from the material it adds. Flip the hem's arrow so it
matches the geometry; edge flanges keep theirs.
A regression test compares the arrow with the side the finished hem adds
material on, for both Flip states, top and bottom edge picks, and both
dialog layouts.
* claim: #3220
* Model cache: keep fewer states of a design too big for its file instead of none (#3220)
A model cache over the container's budget keeps the final state, then the
slowest features' states while the bodies they add fit (a body several
states hold counts once), and is left out only when not even the final
state fits. Opening evaluates just the features whose state the file left
out, each from the state before it, so the model is whole afterwards.
* Model cache: keep each encoded body apart, copy only what a cache stores
Encoding a model too big for its file no longer copies every distinct body
into one buffer before thinning: the payload is copied from the encodings
once its size fits, so saving holds no more than before.
* Model cache tests: canonical runs in the timeline fixture; rustfmt
* Model cache: count the records a thinned cache keeps; always try the final state
Thinning counted only the bodies, so a model of many copies of one part
(records, not bodies, filling the room) shrank its budget to nothing and
saved no cache although its final state fitted. The choice now counts the
records of what it keeps, always keeps the final state, and says what it
counted, so each retry keeps less and the final state alone is always tried.
* docs: a thinned model cache counts its records and always tries the final state
* Model cache test: the slowest feature's state holds bodies 16 to 23
* Model cache: thinned opens with fingerprints that read the state before a feature
Since #3397 a feature's fingerprint reads where the bodies before it lie.
After a state the file left out, adopt takes the saved fingerprint and the
evaluation that works the state out checks it. A result evaluated ahead
is never taken as one the file left out. The thinned round trips include
the placed components' design, whose cut reads those placements.
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
* claim: #4880
* Rename a body: keep the names of siblings a later feature removed (#4880)
Renaming a body sets the stored names of all bodies its feature made, so
the siblings are not renumbered. Those names were read from the final
model, where a body that a later feature removed no longer exists: its
stored name was lost, it came back as `Ext (3)`, and the later Remove
failed with "no body `Body3`". Read the siblings from the feature's own
result instead, as the later features already are.
* Test: renaming one half of a split leaves the other
Renaming one half keeps the other half's name, also once a later feature
joins that other half away, after both an incremental and a full
evaluation.
* claim: #4846
* MCP tools: unknown command ids suggest the closest ids (#4846)
An unknown command id in `execute` or `batch` now lists up to three closest
ids, and an id with parameters packed into it says to pass them in `params`.
* claim: #4804
* Keyboard Shortcuts page: draw only the visible rows (#4804)
Off-screen rows keep their height but skip their cells, the list is
filtered once before measuring and drawing, and each row's shortcut
comes from the command spec already in hand instead of a registry
lookup.
* claim: #4899
* Sketch: offer Break Offset Association for offsets of lines (#4899)
Trim and Extend refuse an associated offset output and point to Break
Offset Association, but the right-click menu offered it only for
free-form offsets. Offer it whenever the curve is an offset output
(analytic, topology or free-form), using the sketch's shared lookup.
* claim: #4851
* solidcraft-cli bench-assistant: the standard Assistant model test
Runs the eleven v1 tasks of #4851 against a desktop app that is already
running with a control port. Each task starts a new design and a new chat
with All permissions, runs only the prescribed setup, sends the prompt
once and waits for the turn. The result is checked on the design
(document.inspect with measurements): body count, volume within 1%
(0.3% for the fillet), bounding-box size in any orientation, the lowest
corner, the body name, live feature kinds, and changed or unchanged
content; the volume question is checked on the reply. Failures get the
issue's categories from the conversation's items. Prints a table, and
writes the issue's comment template (--markdown) and the raw results
with each conversation (--json).
Dropping a feature just behind the history marker now rolls it back, and
dropping it just before the marker makes it active again, in both the
bottom strip and the left panel. timeline.reorder and timeline.move_many
take an optional `active` param that sets the marker in the same command,
so the move is one undo step and one recompute. A position on the wrong
side of the marker for the requested `active` is refused.
Drop feedback:
- bottom strip: the drop line sits centred between two features (it
used to touch the previous one) and shows at the first position
(it used to be drawn outside the strip there)
- left panel: a drop slot below the marker when it is at the end
- no drop line when the marker is dragged back to where it is
- a dragged feature is outlined in the drop line's green
- grab cursor on the marker, grabbing cursor while dragging the marker
or a feature
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4932
* icon_audit: compare the requested-icons manifest with CRLF checkouts ignored (#4932)
On Windows, Git for Windows checks out requested_icons.txt with CRLF line
endings (core.autocrlf; the repo has no .gitattributes), while the test
builds the manifest with \n, so the comparison fails although the lists
are the same. Read the file with \r\n normalised to \n, as #1931 did for
the icon tools. Regenerating still writes \n.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4966
* Agent chat: serve the bundled skills' reference through an MCP tool (#4966)
The bundled skills start with "read reference.md", but SolidCraft's own
chat turns the agent's file tools off, so its agent could not follow them.
The MCP server gains a read-only `skill_reference` tool that returns a
skill's documents from copies built into the binary. It takes a skill
name and a file name from a fixed list, never a path, and refuses an
unknown skill or file with the available ones. The skills point to it,
and docs/mcp.md and skills/README.md describe it.