A model the installed pi-ai catalog does not describe was reported as
text-only with no way to say otherwise, so a vision model added through
the custom-provider form was refused at every image admission point.
The justification in the source described the DeepSeek chat-completions
serializer, which does reject image blocks; the pi-ai request converter
and every wire protocol it speaks carry images.
Modalities now resolve entry `input` -> installed catalog entry -> route
`defaultInput`, the chain the two capacity fallbacks already use, so the
route value is a fallback and never narrows a catalog model. Its default
stays `[text]`: nothing can interrogate a gateway for its modalities, and
over-claiming admits an image the provider rejects mid-turn, after prompt
admission has already committed the message.
The VitePress config declared no position for the subsystem or other-interface
sections, so `indexOf` returned -1 and sorted them ahead of every declared
group: the reference landing page's own sidebar entry sat 1549px below the
fold. Four subsystem pages also shared `order` values with pages in the same
section, resolved only by sort stability and array concatenation order.
Section placement and collapse move into the manifest as a per-locale
declaration, and `sectionSpec` throws for an undeclared section instead of
sorting it silently to the top. Subsystem pages are grouped by concern, the
six topical groups collapse until one holds the page being read, and page
order derives from array position.
The projector drops the language-switcher line and repository badge the
canonical pages carry for their GitHub readers. The navigation bar gains the
DeepSeek wordmark, a release-stage tag, and a favicon; the sidebar scrollbar
rests invisible and appears while scrolling. Subsystem pages carry a two-level
outline, and the two plugin-development tracks now cross-link.
parseCmdline(ctx, program): void only adapts commander control flow to
the launcher: it parses the immutable cmdlineArgs snapshot and turns
help, version, parse errors, and action rejections into a ctx.appExit
request. App validation and the ctx.provide of the app-owned service
live in the program's own synchronous .action(), which commander runs
inside parse — program.error(...) there shares the exit path with a
grammar rejection. Deletes the CmdlinePlan export, its unread ctx
parameter, the type-unsound (() => ({}) as T) default, and the
T | undefined return with its per-caller publish guard.
Overlaps master on the preset e2e, the web scaffold, and the generated
config-catalog triplet; all merged textually, so the catalog is regenerated and
its pairing hashes re-recorded rather than trusted.
Adding COLLAPSE_SECTION_ORDER and CODE_ONLY_INSTRUCTION shifted every
later line in packages/core/tools/src/index.ts by 14, and three generated
artifacts cite those lines: the subsystems cordis-surface region, the
event producer/consumer matrix, and the Agent Note pair record left
inconsistent by the cherry-pick resolution.
`collapses()` read `defaultMode`, so the collapse only applied when the
DEPLOYMENT was `code`. An agent handed `code` by an agent preset under a
native default announced `[run_code]` on the wire and still executed a
model-direct native call -- the bypass this collapse exists to close,
reopened for exactly the composition `dsh-agent-tool-mode` produces.
`modeFor(scope)` is the same resolution `wireSchemas` and the SDK section
already use, so presentation and execution cannot disagree, and a mode
inherited from a standing preset scope collapses like a declared one.
The per-agent and preset tests asserted only the wire, which is why the
regression passed them. They now assert through the executor: the body
never runs, the call resolves UNKNOWN_TOOL, and the native sibling beside
it still executes.
wireSchemas() already advertised only run_code under mode: 'code', but the
executor resolved every call through get(), which returns the full visible
map plus the reserved transport. A model could name a native tool directly
and bypass run_code entirely. Route the execution-path lookups through a
new private resolveExecution() that applies the mode collapse at the
operation boundary: model-direct calls under 'code' may only name run_code
(UNKNOWN_TOOL otherwise), while SDK sub-dispatches (parent token set) keep
every visible tool. get()/schemas() public semantics are unchanged.
The denial happens at createExecution, before the extensible policy
pipeline — pre-execute listeners, approval ask, and guards never observe
a call that is deterministically denied. A collapsed call honors the
pre-dispatch cancellation contract, routes aborted results through the
visible tool's finalizeContent, and captures the finalizer before
argument materialization.
Regenerated docs, catalogs, graphs, scoped events, re-recorded
translation pairs, and updated test assertions.
Fixes#1815
wireSchemas() already advertised only run_code under mode: 'code', but the
executor resolved every call through get(), which returns the full visible
map plus the reserved transport. A model could name a native tool directly
and bypass run_code entirely. Route the execution-path lookups through a
new private resolveExecution() that applies the mode collapse at the
operation boundary: model-direct calls under 'code' may only name run_code
(UNKNOWN_TOOL otherwise), while SDK sub-dispatches (parent token set) keep
every visible tool. get()/schemas() public semantics are unchanged.
The denial happens at createExecution, before the extensible policy
pipeline — pre-execute listeners, approval ask, and guards never observe
a call that is deterministically denied. A collapsed call honors the
pre-dispatch cancellation contract, routes aborted results through the
visible tool's finalizeContent, and captures the finalizer before
argument materialization.
Under code mode, a system-prompt/assemble listener filters out tool:*
guidance sections that told the model to call native tools directly.
The tools:sdk section and SDK types remain so programs can still use
all tools through run_code.
Regenerated docs, catalogs, graphs, scoped events, and re-recorded
translation pairs.
Fixes#1815
Regenerates docs/config-catalog and re-records its pairing hashes: master's
side of that generated file merged textually, which leaves the recorded blob
hashes describing neither side's content.
`Config.roots` defaulted to `[]`, so a deployment that did not patch both roots
in got a roster with none — `dsh run` once booted exactly that and failed
resolving `standard`. The user root never needed an app: `<dshHome>/.agent-presets`
is the same place in every deployment, resolvable here the way
`dsh-skill-local` resolves `<dshHome>/skills`.
The roster now derives that root itself unless `includeUserRoot` is false, and
`apps/cli` supplies only the SHIPPED root, whose path an installed app alone
can resolve. The derived root is appended after every configured root, so a
shipped id still shadows a home directory claiming it and `writableRoot()`
still prefers an explicitly configured `user` root; the set is resolved once at
construction, because a root set that changed between a `list()` and the
`copy()` acting on its answer would author into a directory the caller never
saw.
Every test that pins an exact roster now says `includeUserRoot: false` — the
machine's real harness home must not decide what a golden or an assertion
counts.
