file.sample flow-past: air at 25 mm/s past the upright L-bracket in a box of
2.5 mm lattice cells (Re 83), inlet, outlet, a no-slip floor and slip walls;
the fluid step reports the force on each named face. Its test holds the
drag coefficient on the upright's frontal area between 1 and 3.5 (2.3) and
the lattice's mass change under 1 %. No picture: the viewport does not draw
flow results yet (M12).
Part of #164 (its streamlines wait for a flow display in the viewport)
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
file.sample rubber-seal: a gasket ring of hybrid hex8 elements in four
mapped quarter blocks, squeezed 2.5 mm between plates bonded to it
(nonlinear static, Mooney-Rivlin rubber). Its test frees the plates and
matches the closed form for incompressible uniaxial compression within 3 %,
and bounds the bonded ring's extra stiffness. The plates are prescribed
displacements until implicit contact (#161) can press it into a groove.
Closes#163
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
file.sample drop-test: an L-bracket standing on its foot (a SolidCraft script
in sample.rs) swept into hex8, falling 2 m onto a rigid floor tilted 15 deg;
explicit dynamics for 1 ms with gravity and plasticity. Its test checks the
mass against rho V, the impact speed sqrt(2 g h) and kinetic energy, the
energy balance and the momentum the floor takes. docs/examples.md starts
with it, and its picture is docs/media/examples/drop-test.png.
Closes#162
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
A sample may carry a SolidCraft script (the corpus recipe form) whose design
opens before its commands run, and its script sees the design's bodies.
fieldcraft-engine now builds fieldcraft-geometry with its scripts feature.
The start screen scrolls its sample list, and a test keeps the file.sample
form's choices in step with the samples.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
General contact, on a set's faces or every exterior face of the solids, is mortar contact
with one face list as both sides: each face against every other one that faces it, shares no
node with its element and is not reached through more than half an element, paired both
ways. Tested by a nearly incompressible Mooney-Rivlin C-ring (hybrid hex8) closing on itself
(no penetration beyond the mortar's weak enforcement, mirror symmetry to 1e-6) and, heavy, a
rubber bellows whose folds press on each other.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
An oracle case of uniaxial tension across a tie of non-matching meshes; the tie formulation
in docs/fem.md and the literature survey; ROADMAP M7 and parity updated.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
Each secondary node follows the main surface under it as the weighted average
u_A = sum_J (M_AJ / sum_K M_AK) u_J with M_AJ = int psi_A N_J over the reference overlaps and
psi the secondary faces' dual shape functions (Wohlmuth 2000), so the secondary mortar
matrix is diagonal; the rows join the couplings' constraint elimination in every implicit
analysis. Uncovered secondary nodes stay free with a warning; heat transfer and explicit
dynamics refuse ties by name. Tested by the tie patch test on non-matching hex8/tet4 meshes.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
A tie names a main and a secondary face set and an optional position tolerance; it refers to
its sets, round-trips, is bounded in number and checked for solid faces.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
An oracle case of a block pressed on another and slid with static and kinetic coefficients
(mu_k N carried once slipping), and the docs of the friction formulation; ROADMAP M7 and
parity updated.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
solve.nonlinear counts the slipping nodes, solve.probe gives a secondary node's shear traction
and slip, and the results display draws contact_shear and contact_slip, contact_status coding
stick and slip.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
A return map at the closed secondary nodes on the mortar slip in its rate form (w_A = B_A dx,
B_A = [int Phi_A N_i P, -int Phi_A N_j P]): stick within mu p_A, slip back to the cone, static
coefficient for a node that stuck in the last increment and kinetic for one that slipped; the
friction does the virtual work t . B_A dx, so a slipping interface carries mu N exactly. The
tangent is symmetrised (stick and slip stiffness kept, the coupling to the pressure and the
change of B left out; the residual is exact). Tractions and slip states are committed with the
converged increment. After a chattering cutback, nodes open by less than 1e-3 of their
element's thickness keep their penalty in the tangent for the rest of the step. Results carry
the shear traction, the slip and stick/slip status.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
Two oracle cases: the contact patch test on non-matching meshes in small and finite sliding,
and Hertz's cylinder on a flat under force control (peak pressure, pressure distribution,
subsurface stress). docs/fem.md describes the formulation, the literature survey its papers;
ROADMAP M7 and parity are updated.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
solve.nonlinear summarises each contact (normal force, largest pressure, closed nodes, deepest
penetration, augmentations), solve.probe gives a secondary node's pressure, gap and status per
contact, and the results display draws contact_pressure, contact_gap and contact_status. The
contact commands, the catalogue and the model docs say what nonlinear static steps do with a
pair.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
Frictionless contact pairs enforced by an augmented Lagrangian: nodal pressures from the
mortar weighted gaps, Uzawa updates after each converged increment until the gaps are within
1e-5 of the element thickness, and the consistent tangent (outer products of the weighted
gap gradients and the pressure-weighted Hessians from the automatic differentiation of the
mortar integrals). Finite sliding searches and integrates on the current configuration every
evaluation; without nonlinear geometry the pairs and their mortar integrals are fixed in the
reference configuration (small sliding, linear and exact). The tangent's pattern takes the
face pairs' blocks; the step cuts back when the active set chatters. Initial overclosure is
offset or refused; general contact, 2D models and arc length are refused by name; the linear
analyses refuse contact by name instead of the mesh.
With contact the convergence force scale comes from the converged states, the loads and the
current state. The loads-and-reactions balance check takes the nodal reactions into its scale
(no spurious warning for a part squeezed between held faces).
Tests: the contact patch test on non-matching hex8/tet4 meshes (exact with small sliding,
to 1e-4 with finite sliding), Hertz cylinder on a flat (peak pressure within 3%, half-width
bracketed) with quadratic Newton convergence, overclosure, thread-count determinism and the
refusals; Hertz sphere on a flat (heavy).
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
The overlap of a secondary face with a main face (projection onto the secondary plane,
Sutherland-Hodgman clipping of their triangles, Radon's seven-point rule on the overlap's
fan) and the inverse linear and bilinear maps of both faces, written over a Real scalar so
that f64, Dual (gradient) and Hyper (gradient and packed Hessian, 24 variables) give the
values and the exact linearisation of the segmentation, the normal and the projection.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
An explicit step may ask for mass scaling to a target time increment (on
a set, or the whole model): elements whose stable increment is below it
get the density that reaches it, with their lumped masses, frequency
bound and hourglass scales; adiabatic heating keeps the physical
density. The run reports the elements scaled, the added mass in all and
per section with its fraction, a warning, and the added mass's kinetic
energy in the energy account. Verified on a uniform bar scaled to twice
its increment (added mass 3×, increment doubled, a quarter of the
kinetic energy left to the physical mass). ROADMAP and parity note the
hyperelastic materials and mass scaling.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
Neo-Hookean, Mooney–Rivlin and Yeoh materials in explicit dynamics, in
total form from the deformation gradient (objective by construction):
the one-point hex8 takes its mean F over the undeformed element and the
stress τ/J; tet4 takes the isochoric stress of its own F with the
average nodal pressure of the decoupled volumetric energy (Bonet–Burton
at finite strain). Each element's frequency bound follows the tangent's
largest normal stiffness. Small-displacement steps and thermal strain of
hyperelastic elements are refused by name.
Verified by uniaxial strain to λ = 1.5 of all three laws on hex8 and
tet4 against the material point (to rounding), and the oracle case
explicit-neo-hookean-implicit: a quasi-static explicit pull against the
implicit nonlinear solver (lateral contraction to 0.09%, stress to 0.03%).
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
Its values come from abaqus-docs.mit.edu, which returns 403 Forbidden outside MIT's network, so
they are not from a publicly accessible source. The section is kept aside for review.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
A README section tabulates, per benchmark, the reference, the result the
Abaqus 2017 Benchmarks Guide publishes (element, mesh, error) and ours from
the oracle, with a cited source per value and notes where the rows are not
like for like. Linked from docs/verification.md.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
arc-shallow-arch: a sinusoidal arch of 40 × 2 × 1 hex20 under a sinusoidal dead load scaled to
the one-mode closed form's upper limit load: the upper limit point at 0.992 (−0.8 %), the lower
at 0.219 (0.8 % of the peak load) and exactly one negative eigenvalue between them. ROADMAP and
parity updated, docs/oracle.md regenerated.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
NonlinearStatic.arc_length {constraint, max_load_factor, displacement_limit}: Riks–Crisfield
arc-length increments (cylindrical or spherical arc, tangent predictor signed by the angle
criterion, the constraint's quadratic root chosen by the angle with the last iterate), arcs
grown, shrunk and cut back as increments are. The converged state's tangent is factored for the
next predictor (LDLᵀ when indefinite); its negative eigenvalues are recorded per increment and
changes in their count are reported as limit or bifurcation points. Ends at the maximum load
factor, a displacement limit or the increment cap. Abaqus *STATIC, RIKS; LS-DYNA refuses it by
name; forms gain the arc-length group. A shallow arch of hex20 snaps through on the closed-form
path (1.9 %), both limit points found, inertia 0 → 1 → 0 between them.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
Fixes the build of develop: #171 crossed with the typed command forms, so BUCKLE lacked its
form. SOLVE_BUCKLE follows SOLVE_FREQUENCY; the step form gains the buckle variant and the
nonlinear step's imperfection group.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
NonlinearStatic.imperfection {step, mode, scale}: the buckling step's mode, scaled to a largest
component of `scale`, is added to the node coordinates before the step is solved, with a warning
naming it; Abaqus and LS-DYNA decks refuse it by name. A clamped–free column with its first mode
as imperfection under half its buckling load amplifies it by 1/(1 − α) within 2.8 %.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
buckle-euler-pinned and buckle-euler-cantilever (hex20 columns, −0.18 % from Euler with
Engesser's shear correction) and buckle-plate-square (a simply supported square plate, b/t =
100, −0.39 % from k = 4; the support and thickness choices in its README). ROADMAP and parity
updated, docs/oracle.md regenerated.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
Procedure::Buckle {modes, shift}: the step's loads as a linear preload, K_σ from its stresses,
and (K + λ K_σ) φ = 0 by block Lanczos in the K inner product: K⁻¹(−K_σ) on Cholesky without a
shift, the buckling transformation (K + σ K_σ)⁻¹ K on LDLᵀ with one, whose inertia counts the
load factors below σ (Grimes–Lewis–Simon 1994). Lanczos gains an operand and an eigenvalue map
(defaults keep frequency steps bit for bit) and reports converged negative eigenvalues. Results
are frequency results with a `buckling` part; modes display with their load factor.
solve.buckle (in the Static drop-down); Abaqus *BUCKLE, Nastran SOL 105 and LS-DYNA refuse it by
name. Euler columns of hex20 within 0.18 % of Engesser's load.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
K_σ = ∫ (∇N_a · σ ∇N_b) I dV for a stress state at the stiffness points (solids, plane and
axisymmetric elements with the hoop term), for linear buckling. Checked: uᵀK_σu = σ V for a
uniform stress on a distorted hex20 and a plane strain quad8, rigid translations in its null space.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
CommandSpec::listing gives id, label, the parameter doc and the schema; MCP
list_commands, the control channel's engine.commands and the new
fieldcraft-cli commands --json use it.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
The context panel's form shows typed fields instead of the JSON text: numbers
with the model's units, drop-downs of choices and of the model's sets,
materials, steps and bodies, checkboxes, vectors, groups and one-of variants
(a load's type switches its fields), lists of rows. Defaults are filled in,
required fields are marked, a value the schema finds wrong is noted under its
field, and OK waits while a field holds text that is not a value. The JSON
text stays as the Advanced view. fields.rs holds the generic widgets (row,
number, drop-down, text), adapted from SolidCraft's dialog rows.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
cmd/forms.rs describes each command's parameters: model definitions (sets,
materials with elasticity, plasticity, hyperelasticity, damage, thermal and fluid
options, sections, reference points, couplings, masses, springs, boundary
conditions, loads, fluid domains, steps including nonlinear and explicit),
meshers, solvers, deck export, files and results. Tests check that every command
has a form and that a form's defaults are accepted or refused only for the
required fields left empty.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
Field, Kind, Dim and Collection describe a command's JSON parameters: numbers with
unit dimensions, choices, vectors, references to model items, groups, one-of
variants and rows. prefill, switch_variant and problems serve forms; json_schema
gives the same as JSON Schema. No command is described yet (form: None).
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Segment-to-segment penalty contact between element faces: each secondary
face and each main face facing it are projected onto the secondary face's
plane, split into triangles, clipped (Sutherland–Hodgman) and integrated
with a three-point rule per sub-triangle (Puso–Laursen segmentation, the
penalty at the integration points as Fischer–Wriggers), so a uniform
pressure gives the consistent nodal forces on both meshes and the contact
patch test passes on non-matching meshes. Every contact runs both ways at
half the penalty: pairs with each surface as the secondary, general and
self contact face against face. A bucket grid of face boxes every
increment, faces integrated in parallel and their forces added in face
order: the same numbers on any thread count.
The penalty per area ε = s γ min(m/A)/Δt² comes from the lumped masses,
each node's directional area and the stable increment, bounded by
Gershgorin's row sums so contact never shortens the increment. Coulomb
friction with exponential decay by a regularised return map at the
secondary nodes; rigid walls take Coulomb impulses. Contact work and
friction dissipation join the energy balance; each contact reports its
peak force, first and last contact, deepest penetration, friction and the
nodes adjusted at the start (initial penetrations adjusted strain-free or
refused with the node, face and depth); a cap on contact faces.
Verified (crates/explicit/tests/contact.rs, tests/oracle/explicit-*):
colliding bars (2L/c +1.2%, ρcAv₀/2, KE kept to 0.005%, contact energy
back to 8e-5 of it), blocks sliding to a stop on a rigid wall and on a
plate under constant and decaying friction (0.005–0.2%), the patch test
(σ_zz uniform to 0.06%), two spheres colliding against Hertz (peak +2.3%,
duration −0.4%; the oracle case waits for a graded sphere mesher), a free
body untouched by general contact, determinism across threads. Oracle
measures gain `over` (ratios). Existing fixtures do not move.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
Contact definitions on face sets: a pair (each surface's nodes against the
other's faces), general contact on a surface (self contact) or on every
exterior face of the solids; Coulomb friction with static and kinetic
coefficients and an exponential decay, a penalty scale and the choice to
adjust or refuse initially penetrating nodes, validated with caps. Rigid
walls of explicit steps gain friction. The check wants solid faces on
contact surfaces; sets used by a contact cannot be deleted.
fem::contact: the closest point of a triangle (Ericson) or a bilinear
quadrilateral (projected Newton) to a point, and a bucket grid of face
boxes for the global search, deterministic and bounded.
Commands interaction.contact.pair, .general, .edit, .delete and .find
(the touching parts of the mesh, with face sets and pairs made of them);
contacts in the model browser. Abaqus writes *SURFACE INTERACTION with
*FRICTION (exponential decay), *CONTACT PAIR and general *CONTACT; LS-DYNA
*CONTACT_AUTOMATIC_SURFACE_TO_SURFACE and _SINGLE_SURFACE with FS, FD, DC;
Nastran refuses contact by name. The implicit solver refuses contact by
name (M7); the explicit solver refuses it until the next commit.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
nl-hybrid-incompressible: 27 hybrid hex8 at ν = 0.4999 pulled to λ = 2 by a prescribed
displacement, against the compressible neo-Hookean closed form (1.4·10⁻⁹); its first
iterations need the LDLᵀ factorization. ROADMAP and parity updated, docs/oracle.md regenerated.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
A tangent that is not positive definite is factored by LDLᵀ (Cholesky first, one symbolic
factorization) instead of cutting the increment back; each iteration records the tangent's
negative eigenvalues and the summary warns when a converged state keeps some. A model that can
move freely is still an error at its first tangent, located by the undeformed Cholesky factor;
a singular tangent later is a cutback. Nearly incompressible hybrid hex8, hex20 and tet10 at
ν = 0.4999 under prescribed displacement now converge to the closed form.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
A supernodal LDLᵀ on the Cholesky factor's symbolic factorization (ordering, supernodes, fill,
cap and parallel subtree tasks shared through a Symbolic and a generic driver; the Cholesky
numerics are unchanged bit for bit). Pivots by the bounded Bunch–Kaufman (rook) search of
Ashcraft, Grimes & Lewis among each supernode's columns (Schenk & Gärtner's supernode
Bunch–Kaufman), 1 × 1 and 2 × 2, right-looking panels whose trailing update reuses the tiled
Cholesky kernels with L D formed on the fly. Reports the inertia; a column without a pivot above
round-off is a located Singular error; NaN is refused. Symmetric::new picks Cholesky, LDLᵀ, or
Cholesky with an LDLᵀ fallback; solve_refined adds iterative refinement. 1.1–1.2× the Cholesky
time on a 140 000-equation hex8 pattern.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
The owner's final choice: the hex-meshed cube with its banded fringe,
drawn a little larger, with no streamlines at any size. The Icon Composer
document is back to the floor and cube groups.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The streamlines are now banded in the fringe colours along their length, as
flow speed would be, with a thin dark halo and arrowheads. Two sweep in from
the lower left, climb over the cube and run straight into the top face's
mesh lines, narrowing to them, so the flow becomes the structure; one wraps
around the cube. The small variant keeps the cube with one bold coloured
streamline and one arrow.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The mark is now a hex-meshed cube in isometric view whose faces carry a
banded colour fringe (blue to red, hottest at the front corner), with three
tapered streamlines wrapping around it. The small variant keeps the fringe
cube and one bolder streamline. Icon Composer gets four groups (floor, flow
behind, cube with mesh, flow in front), its maximum. The title-bar mark is
redrawn as the fringe cube.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
pages.yml builds apps/fieldcraft-web (trunk build --release, wasm-opt)
on pushes to main and develop and on manual runs, and deploys the site
to https://jyviko.github.io/fieldcraft/ only from main (releases).
Develop merges that touch code build without deploying, so a broken
web build shows up before a release. Trunk is a pinned release binary
checked against its SHA-256; actions are pinned by commit.
Part of #119
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
The desktop's ModelApp in a web page through eframe's web runner
(WebGPU, WebGL2 fallback, `?webgl` to force it), built with Trunk.
The browser services sit behind the UI's host-service seams: File >
Open uses the browser's file picker, dropped files are read
asynchronously, and both go into the page's files; Save, Save As and
Write Deck hand the written file out as a download. Autosaves go to the
recovery folder as on the desktop and are mirrored to local storage, so
File > Recover offers what a closed or crashed page left. `?sample=NAME`
opens a sample. A loading screen with the app icon, and a message
instead of the app when neither WebGPU nor WebGL2 is available.
`cargo xtask wasm` (and so `cargo xtask ci`) now checks fieldcraft-web
for wasm32 with the library crates. docs/web.md: running it locally,
what works in the browser and what does not.
Part of #119
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
The browser build has no threads, so a background job (opening a file,
meshing from a form, solving) ran inline in the click and the page froze
before it said what it was doing. Now the work waits until the window
has drawn the job's progress, then runs on the window thread in the
next frame. Native jobs still start on a worker thread at once.
Part of #119
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
