Semantic resolutions beyond line merges:
- The bash seam keeps resolveMode + the bash/resolve-mode waterfall on
master's task-free BashExecutor (run/start/resolve only; BashProcess
handles); tool-bash consults it at its stamping site and escalation
baseline on master's render/background split, with a waterfall test on
the recording executor.
- dsh-mode's BASH_FAMILY narrows to ['bash']: bash_output/bash_kill are
replaced by the kind-generic task_output/task_kill, which span every
task kind and only observe or stop work, so the access cap withholds
only the starter it can reason about.
- The plan-mode snapshot suite adopts master's pin grammar (tool-schema
sidecars; the expectedHeaderSnapshots extension is gone — the exit
transition deltas, and entering-before-turn-1 needs no second
snapshot); modes-advertise joins the plan header class (no-model, so
membership is vacuous). Fixtures re-recorded on the acp-demo bin;
the replay overlay gains the passthrough sandbox runner.
- examples/plan-acp-agent rewires to @deepseek-ai/dsh-acp-demo and drops
its tool-bash entry (the spine bundle now composes it); dsh-stdio (the
renamed stdio-chat home) keeps its /mode command and gains the dsh-mode
peer edge; the acp bridge keeps the modes surface beside master's
permission presets.
- mode README gains the Model Experience / Known Limitations sections the
new README gates require; AGENTS.md ceiling 1370 → 1440 for the kept
mode/ layout line and Agent efficiency section.
The background-task change repeated its lifecycle design across implemented RFCs, package READMEs, JSDoc, test commentary, and model-visible schemas. That repetition obscured the contracts that maintainers must preserve and added avoidable prompt tokens.
Rewrite the implemented RFCs around the current design, keep authorization, exact-owner cleanup, wait/abort ordering, producer quiescence, and teardown-failure guarantees at their owning surfaces, and remove peer surveys, review history, control-flow narration, and emphatic restatement.
Shorten the task and subagent schema wording, synchronize the bilingual tool cookbook, and regenerate the config, service, RFC, tool, and replay snapshot derivatives. Runtime behavior is unchanged; test edits update prose-only assertions and descriptions.
Conflict resolution beyond line merges: the ACP bridge's mode/set
re-notify moved inside master's new containment finally (a throwing
presenter must not desync the picker), and session/new keeps the modes
advertisement beside master's plain object return. dsh-mode's boundary
flush moved off the session/event feed onto the loop's interception
seams (agent/prompt-submit / agent/turn-continuation): master's scoped
dispatch made post-commit session observers observe-only — an append
from one re-enters the publishing append — while the two waterfalls
fire outside any log emit and bracket the same boundaries; mode.spec's
boundary helper now dispatches those seams via agentEvents, and the
containment tests stub only the mode/set append. Catalogs, doc graphs,
and the module graph regenerated.
A ModeDefinition is now exactly { section, access? }; unknown keys (a
tools list included) fail loud at load. What plan mode still does: the
guidance section, the exit_plan_mode visibility rule (plan only, both
soft surfaces), the access cap's bash/resolve-mode clamp, and the two
cap-derived pre-execute guards (the bash trio is withheld when no
confining executor can honor the cap; sandbox escalation is denied
while it holds). The general deny-by-default gate and the assemble
allowlist filter are gone: which tools a mode admits is an effects
question, and a hand-maintained name list mislabels it — it must track
every composed tool and rots silently as tools arrive. The dimension
returns as a consumer of effects self-declaration on tool definitions
(MCP ToolAnnotations as the template) — rationale and restart trigger
archived in the RFC's Alternatives/Deferred; the interim guidance-only
non-shell restraint is priced in Consequences.
Exiting plan is now a pure removal (the exit tool + section), which the
delta encoding CAN express: the re-recorded plan-mode fixture pins one
plan-shaped initial header snapshot plus one header-delta instead of
two snapshots.
The stack rebases onto a moved master through its base branch. Beyond
mechanical unions (both branches' demo scripts, example rows, service
roles, tool lists, acp deps, doc budgets — each side fit alone, the
union needs the higher ceilings), three semantic reconciliations:
- The ACP bridge now carries BOTH per-session surfaces: the sandbox
stack's config options + approval answerer and this branch's session
modes; session/new and session/load advertise modes AND configOptions
side by side.
- The feature matrix supersedes the sandbox stance per the RFC's
second-lander rule: session/set_mode and current_mode_update flip to
shipped-by-dsh-mode, config-option rows stay as #169 wrote them, and
§6 records both landed features under the picker-to-modes /
knobs-to-config-options division.
- The snapshot pin grammar (#254: one header snapshot + declared deltas
+ a Markdown prompt golden) gains a symmetric declaration for what a
delta cannot express: expectedHeaderSnapshots — a plan-mode flip
resorts the canonical tool list, so its widening lands as a second
full snapshot, now its own Markdown section. The pin-less-class and
model-turn-only-pin amendments carry over; new fixtures cover the
extended writer paths, and the plan-acp-agent scenarios re-recorded
under the merged composition (the app now bundles the skill tool)
with the suite's refresh mode wired through.
dsh-bash grows the per-call policy carrier: BashExecRequest.sandboxMode
(request-optional, spec required-but-nullable — the owner pattern; resolve()
is the one explicit defaulting step) and the BashExecutor.sandboxMode
capability fact (undefined in the base class — composition truth the tool
layer can read). dsh-bash-local carries the field verbatim and confines
nothing.
dsh-bash-sandbox extends LocalBashExecutor and hands ctx.sandbox the exact
argv it is about to spawn. A denial is a RESULT FACT (the command RAN;
result.sandbox.denied is orthogonal to exitCode/signal), classified
conservatively against the wrap own dialect; a RUNNER failure outranks
denial — foreground re-throws the structured SANDBOX_UNAVAILABLE, a settled
background task stamps sandbox.runnerFailed — so a broken sandbox never
reads as a failing command and the command never runs unconfined.
dsh-tool-bash renders the markers and teaches the model not to retry around
a policy denial; escalation and per-session switching are staged follow-ups.
Review finding: the exit tool's direct mode/set append flipped the
folded mode while the loop could still execute further tool calls from
the SAME assistant response — a same-batch exit_plan_mode + write pair
would sail past tools/pre-execute under 'default' even though the
request was assembled under the plan-shaped header. That broke the
design's own invariant (a step's executions run under the mode its
assembly folded), which the pending-intent flush was built to hold for
user flips.
The tool now records the switch as a pending intent like every other
writer, flushed at this step's end (still in-turn); pending intents
carry a narrate flag so the exit's flush stays silent — the tool result
is its narration — while user flips keep the coalesced boundary notice.
The gate, folding the logged mode only, now provably covers the whole
batch: regression test pins approve-then-write-in-the-same-batch as
denied, and the widened toolset still arrives on the next step.
The recorded scenarios are re-recorded: the fixture now shows mode/set
landing after step/end, before the widened fallback header.
Plan mode's stage 2 (RFC 2026-07-07-plan-mode). The exit tool: one
required plan argument (the durable log artifact), execute re-checks the
folded mode, then conducts the review over the user-interaction seam —
one single-select question (Approve / Keep planning) with free text open
— so an approval appends mode/set back to default in-turn and every
other outcome (keep-planning feedback verbatim, aborted, no provider)
returns the corrective isError with the mode unchanged. presentCall is a
generic card titled by the plan's first heading carrying the plan
markdown; over ACP the review rides the ask_user elicitation flow, in
the terminal the stdio prompt queue — no approval-seam dependency.
The ACP bridge maps the picker 1:1 onto ctx.modes (opportunistic, a
type-only peer edge): session/new + session/load advertise
availableModes/currentModeId, session/set_mode validates through set()
and echoes an optimistic current_mode_update (the pending mode IS the
selection; the logged mode/set lands at the boundary and, matching, is
not re-sent), and a session/event listener re-notifies on each logged
flip that differs from the last sent — the tool-driven exit updates the
picker. The feature matrix rows move from 'not modeled' to the
picker-to-modes / knobs-to-config-options division, with the ACP v2
removal direction recorded as a mechanical-migration risk.
The snapshot harness gains the setMode/setModeExpectError ops and a
scripted elicitationAnswers FIFO (cancel on exhaustion; a stray choice
string reaches the agent verbatim as a non-consenting custom answer, so
a scenario bug fails safe). The suite factory's header-pin requirement
now applies only to model-turn scenarios — a protocol-only suite has no
header content to anchor. examples/plan-acp-agent is the live
composition; its keyless modes-advertise scenario pins the wire surface
(advertisement, both set_mode round-trips, unknown-id rejection). The
recorded plan-mode approve/reject arc awaits a with-key recording
session; its texts are pinned at the unit tier meanwhile.
examples/AGENTS.md ceiling 653 → 680: the new example's required smoke
row does not fit the old budget.
Adopts #211 (Code Mode tools: run_code + the code/both-mode snapshot
scenarios). Tool-catalog expectations take the union (run_code +
task_*); the two new pinsHeader fixtures (code-mode-turn,
both-mode-turn) were recorded on master without the task runtime, so
they are re-pinned KEYLESSLY by replaying their recorded chunks against
the merged tree (same procedure as text-turn) — the fixture diff is
exactly the header delta: task tool schemas, the tool:tasks prompt
section, and the bash background wording.
Adopts #220 (tool-cordis + the gen-cordis-api gate: the runtime API
catalog regenerated with ctx.tasks/onCleanup on it) and #225 (shared
transcript renderer). Tool-catalog expectations take the union
(cordis_* + task_*); packages/README budget adopts master's 660
ceiling, which absorbs both new group rows.
start({ kind, label, owner, run }) preflights everything that can fail
(the attachSurface fence, validation, the owner-cleanup attach) BEFORE
invoking the producer's run() starter, then commits atomically —
'work started but never got a collectable id' is now structurally
impossible instead of a producer try/catch rollback obligation (the
P1 review fix, rebuilt on #185's declare/execute split). Producers
lose their catch-wraps; the leak tests now pin the stronger property
that a failed preflight never spawns anything. TaskRegistration splits
into TaskStart (identity + run) and TaskHooks (cancel/done/readOutput);
docs, type-equiv manifest, catalogs, and both RFCs move with it.
Adopts #185 (dsh-timeout: clampTimeout/deadline/timeoutOf drive bash
run() timeout classification; runBash loses its own timer) and #108
(ask_user_question) across the task-runtime rework: bash-local keeps
the BashProcess handle shape with master's deadline mechanics, tool
catalogs/expectations carry both the task_* and ask-user tools, and
generated docs are regenerated on the union.
One shared ctx.tasks registry (branded <kind>-N ids, owner-fenced
read/kill/wait/list, attachSurface misconfiguration fence, reported-flag
notice dedup, atomic register) + dsh-tool-tasks (task_output/task_list/
task_kill, completion-notice injection, background prompt habit).
Producers opt in via their own enableRunInBackground config: bash
(stream kind; seam slimmed to resolve/run/start returning a BashProcess
handle, bash_output/bash_kill deleted) and subagent (final-output kind;
done settles after run.dispose()). Owner disposal drains tasks through
the new awaited ctx.agents.onCleanup seam in the loop's disposal chain.
Both RFCs moved to implemented/; docs, catalogs, snapshots re-pinned.
Implements docs/rfc/implemented/feature/2026-07-09-bash-backed-grep-glob-
discovery.md: model-facing glob/grep in a new @deepseek-ai/dsh-tool-fs-search
package, executing fixed ripgrep templates through ctx.bash.resolve/run —
not ctx.fs provider methods — so filesystem backends stay free of a search
contract and sandboxed/remote executors substitute cleanly. The tools never
call ctx.bash.start(); the tool layer owns quoting (one singleQuote safety
boundary), rg --json parsing, ItemRetainer/TextRetainer retention, and the
first tool-owned ctx.spillFiles.saveText() handoff (item-level retention the
generic post-execute spill policy cannot recover).
RFC amendments on the way to implemented/: a shared src/search-core.ts (the
SEARCH_* vocabulary + bash-run/raw-spill/spill plumbing was byte-identical
across both tools — the missed-extraction smell), and a snapshot-gap note:
wiring the acp-agent tree changes the assembled prompt, so goldens need a
keyed re-record; the spill notice text is pinned by unit tests instead and
only the coding-agent example ships the tools for now.
P1 review finding: extractMeta timed only the literal's vm evaluation;
materializing the RESULT then read properties ordinarily on the HOST
stack, so a meta literal smuggling a getter (get name() { while(true){} })
could wedge the host outside any timeout — defeating the exact spin
isolation the worker thread exists for.
Rather than harden the evaluator (descriptor walks, AST validation),
delete the mechanism: the workflow's identity now reaches the seam as a
plain JSON field (WorkflowStartRequest.meta), carried by the tool as a
schema-validated `meta` object parameter the model fills directly. The
engine only shape-validates data (validateMeta, every violation named)
and pre-parses the body; the scanner, the vm evaluation, and the
host-side materialization are gone, and with them the hole. A body
still opening with a Claude Code-style `export const meta` statement
gets a pointed SCRIPT_PARSE message (the likeliest authoring slip; a
CC script's body stays drop-in, only its meta header moves into the
parameter). syncTimeoutMs now governs exactly one thing: the initial
synchronous slice inside the worker.
The RFC's decision section is rewritten in place (implemented-RFC
rule); the embedded-meta format moves to alternatives-considered with
the hole as the reason. Tool description, presentation (title now reads
meta.name directly — the textual sniff is gone), seam vocabulary docs,
and catalogs follow.
Two review findings (#220) on the sandbox context façade:
- Undeclared services were reachable: the façade resolved any live global via
ctx.get(name), so ctx.bash worked without inject: ['bash']. A cross-mount
consumer could then depend on a provider cordis never saw — unmounting the
provider would neither park the consumer nor unwind its registered tools,
leaving a model-visible tool that fails only at execution. The façade now
reads ctx.fiber.inject and refuses any service the mount did not declare
(with a teaching error naming the inject fix), so the dependency is always
visible to cordis and its activation/unload semantics bind.
- ctx.tools.get returned the live ToolDefinition, including execute — mount
code could call another tool directly and bypass ToolRegistry.execute and
its pre/post-execute hooks and accounting. get now returns the same
read-only name/description/parameters view as schemas(), never an invocable.
Adds inject-gate and schema-view regression cases to sandbox-context.spec.ts
(undeclared property/get denied, declared allowed, the cross-mount zombie-tool
scenario refused at call time, get exposes no execute). Package stays at
per-file 100% coverage. RFC, mount description, and tool-catalog updated.
Review finding (#220): the guarded proxy only special-cased ctx.tools, so
mount code could reach an UNGUARDED context through ctx.root, ctx.extend(), or
a service instance's .ctx, then ctx.root.tools.register({…}) to bypass the
marker check and host-realm normalization — a raw vm-realm result would later
error a real agent turn at the session-log plainness check.
The sandbox ctx is now a whitelist façade, not a pass-through proxy: it exposes
only what a mount needs — tools.register (marker-guarded), on/once, provide, the
timer helpers, and injected services resolved through a guarded get — and denies
every framework-plumbing member (root, parent, fiber, reflect, registry, extend,
isolate, intercept, plugin, set, mixin, …) with a teaching error. Injected
services are wrapped so a method returning a Context is rejected on the way back
(the .ctx escape), closing the one indirect leak. There is no context-valued
member left to reach; cross-mount provide/inject is untouched (the plugin's own
inject and the fiber's pending/active gating are unchanged). ctx.plugin (child
plugins) and ctx.set are denied by design; ctx.effect is deferred (FIXME).
Adds tests/sandbox-context.spec.ts covering the escape class (root/extend/fiber/
plugin/set/… denied, the classic root.tools.register bypass, the .ctx escape,
read-only writes) plus the async-service and symbol/in-operator paths for 100%
coverage. RFC/README/tool-catalog/config-catalog updated; api-catalog.ts
regenerated (also picks up the codeRuntime service that entered on the master
merge and was left stale).
The tree SHAPE was the wrong surface for the model: what it needs from
cordis_inspect is what services, plugins, and capabilities are loaded, not the
fiber hierarchy. The plugins section is now a flat name + lifecycle-state list
from ctx.registry (deterministically sorted, one line per instance); the ASCII
tree renderer, the parent→child rebuild, and the dyn-id tree annotation are
deleted — dynamic mounts keep their own richer dynamic section (id, state,
provides, waits). Net -49 lines; RFC and READMEs state the flat-list contract.
Field sessions showed models writing tool schemas in the JSON-Schema dialect
by strong prior — type: 'integer', required: false, then the full
{ type:'object', properties, required: [...] } wrapper — and the rejection
text itself pushed a nearly-correct DSL attempt BACK to raw JSON Schema: one
stats tool cost three consecutive schema errors before mounting. The boundary
now normalizes wherever the input has exactly one meaning (wrapper unwrapped
with the required array becoming per-property flags at any nesting level,
integer → number, required: false → optional, all rebuilt as fresh host-realm
objects) and rejects only genuinely meaningless input, enumerating the valid
vocabulary in the error. Re-running the failing session mounts first-try.
The mount description documents both accepted forms.
The sandbox deliberately provides no Node API, and now says so instead of
letting a bare ReferenceError teach nothing: require, the timers, and fetch
are callable traps whose error redirects to the cordis alternative (inject:
['fs'] + ctx.fs, ['web'] + ctx.web, ['bash'] + ctx.bash, ['timer'] +
ctx.setTimeout — a fiber effect, unwound on unmount). Only function-shaped
globals are trapped; process/Buffer stay undefined so typeof feature probes
stay inert. The mount description and the demo persona state the routing rule,
and the demo mounts ctx.fs (local provider) and ctx.web (seam + keyless local
fetch provider) so agent-built plugins have real capabilities to build on.
Live-validated: a model that reached for Node setTimeout self-corrected to
inject: ['timer'] in one step and built a working ctx.web fetch tool.
New top-level packages/cordis/ group with the self-referential toolset:
cordis_inspect (services / plugin tree / tools / dynamic mounts / api / events,
the api section intersecting the generated catalog with the live service store),
cordis_mount (model-written code evaluated in a node:vm sandbox, mounted under
one cordis-dynamic group fiber as dyn-<n>), cordis_unmount (awaited disposal to
quiescence). Boundary mechanisms: dual-realm instanceof, JSON realm
normalization of dynamic tool results, marker-guarded registration, SchemaSpec
teaching errors, parse failures surfaced with the offending line + caret and a
line-scoped TypeScript hint, and the unmount-first recipe on tool-name
collisions. Config: vmTimeoutMs (schemastery, default 5000). Design record:
docs/rfc/implemented/feature/2026-07-08-self-referential-cordis-toolset.md.
The tool-catalog boot manifest, its regenerated output, and the pinned
tool-name list land here rather than with the other repo registration: the
completeness guard globs packages/*/tool-* and fails the generator (and the
core/tools spec) the moment the package directory exists.
Tianyi Cui
Yichen Jiang
Dudu-0223
kingwl
imccyu