enableRunInBackground: false removed the parameter from the advertised
schema only — the arg validator deliberately allows undeclared keys, so
a caller (or a model that has seen the parameter elsewhere) could still
force run_in_background: true and start background work past the
deployment's opt-out, in both tool-bash and tool-subagent. Both
producers now refuse the forced key loud in execute(); tests pin the
refusal (and that nothing spawns) alongside the untouched foreground
path; the schema-omission-is-advertising rule is recorded in the
runtime RFC and both READMEs.
Two windows where settle() could read a live waiter count, mark the
task reported (suppressing the completion notice), and then watch that
waiter reject with 'wait aborted' delivering nothing — leaving the
owning session with no terminal notification at all:
- abort and settlement in the same tick, settle continuation ordered
first: the waiter now un-counts itself SYNCHRONOUSLY inside onAbort
(the finally decrement alone lands a microtask too late), so the
settle path sees no live waiter and the notice fires;
- abort landing after settlement but before the wait's resolve
microtask: the wait now resolves and DELIVERS the terminal snapshot
it owes instead of rejecting (settlement suppressed the notice on
this waiter's behalf).
Both windows pinned by deterministic tests that fail on the previous
implementation; wait()'s abort contract updated in JSDoc/README/RFC.
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.
The sandbox execute wrapper JSON round-tripped the return and blindly cast it
to ToolExecuteReturn. A JSON-valid but wrong-shape return — a bare string,
{ content: 'ok' }, blocks without a type tag — sailed through: the registry
spreads result.content, so { content: 'ok' } became ['o','k'], passed the
session log's isJsonValue gate, and the DeepSeek serializer then flattened it
to '(no output)' — silent corruption of the next model request and every
replay, instead of a contained tool error.
The round-tripped value is now shape-checked against the two ToolExecuteReturn
forms (array of content blocks, or { content: blocks, meta? }); block checks
are structural only (plain object + string type tag) because the ContentBlock
union is merge-extensible. A wrong shape — and the formerly cryptic
forgot-return/bare-string cases — fails that one call with a teaching error
echoing a truncated preview of what was returned and the two valid forms.
New specs pin the object-form pass-through (meta included), six rejection
shapes, and the preview truncation; per-file 100% coverage holds.
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.
ctx.tasks.wait arms a deadline() fusing the caller's abort with the
wait timeout and classifies the outcome with timeoutOf scoped to the
new TASK_WAIT_TIMEOUT code: a wait timeout resolves to the live
snapshot (the task keeps running), a caller abort rejects the wait —
same contract, no hand-rolled timer/listener plumbing, and a nested
foreign deadline can no longer misread as a wait timeout. task_output
deliberately declares NO ToolDefinition.timeoutMs: timeout-policy
turns a timed-out call into a structured TOOL_TIMEOUT failure, but a
timed-out wait is a SUCCESS that must still report [status: running]
(decision recorded in the runtime RFC alternatives).
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.
The sandbox docs overclaimed a containment contract the design never makes:
"capability access is routed through cordis services, never Node built-ins,
so everything a mounted plugin does stays inspectable and disposable". The
host-realm helpers on the sandbox global (harness, console, btoa) are
reachable functions, so mount code that goes looking can reach the host realm
through one of them — accepted under the trust stance, because the ctx a
mount ultimately receives is fully privileged anyway. Reword the sandbox
module doc, the README trust stance, and the RFC sandbox-semantics section to
say exactly that: the traps and small global surface STEER honest code onto
the cordis services; they are not a security boundary.
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 design record for tool-cordis: the three-tool contract, the vm sandbox
trust stance and boundary mechanisms, the dynamic-group lifecycle, cross-mount
provide/inject composition, the generated runtime API catalog, and the
alternatives weighed (per-capability registration tools, hand-maintained API
tables, a mount provenance event, a hardened sandbox).
The seam shipped without its own RFC — the reconstructable-requests RFC
was amended with the mechanics, but the decision record (why a
compose-once frozen prefix, and what the per-request before/after shape,
a system-prompt section, a durable history opener, per-turn composition,
and a dedicated session event each lost to) had no home. Implemented
lifecycle, feature class, dated to the first commit of the work.
ds-review-bot critical (follow-up): on the first step of a resumed or
seeded/forked instance, auto-compaction ran before runStep composed
this instance's prefix, so the gate read the PREVIOUS instance's logged
prefix from the header fold — a contributor that grew across
resume/fork (skills added, AGENTS.md grown: exactly the
environment-dependent case) could under-gate and ship an over-window
first request.
The loop now composes agent/session-prefix before the instance's first
agent/pre-step (still once per instance; runStep just reads the cache),
and agent/pre-step carries the composed prefix to its listeners.
CompactService.compactIfNeeded gains the sessionPrefix parameter;
BasicCompactService.estimatePressure gates on the handed value — the
header-fold read is gone, so the estimate is exact at every step
including a resumed/forked instance's first. Composition moving before
the boundary snapshot also means a composing listener's session append
now joins the CURRENT request (documented on the seam).
New coverage: composition precedes pre-step and the seam receives the
composed prefix; cancel and disposal landing inside the composition
window drop the step cleanly; the compact gate test hands the prefix
directly.
Review discussion converged on the industry shape (Claude Code caches
user context per conversation; Codex separates initial context from
diffs; Kimi appends at continuation boundaries to protect prompt
caching): stable openers belong in a compose-once prefix, mid-session
changes belong in append-only history — not in a per-request slot.
agent/session-prefix fires ONCE per loop instance, lazily on its first
request-building step: the composed Message[] is deep-frozen, cached on
the transmission bookkeeping, recorded as EpochHeader.messagePrefix on
the anchoring 'initial'/'resume' snapshot, and reused verbatim for
every request the instance sends — prefix stability is structural, not
a producer discipline, and a resume recomposes with attributable drift.
The request is messagePrefix + boundary snapshot.
The per-step RequestAdvice/RequestAdviceContext surface and the
messageSuffix header field are dropped: the tail slot had no consumer,
and every current update pattern (new AGENTS.md discovered, memory
update, skills change) routes through the existing append-only history
channels — inject(), tools/post-execute additionalContext,
prompt-submit additionalContext — each paid once and prefix-cached
thereafter. The messagePrefix delta arm stays for codec totality; the
loop never produces one in practice.
The RFC's deployment-policy decision is unchanged; state the current
mechanism in place — the per-tool budget is declared on ToolDefinition
(timeoutMs, set by the owning tool plugin from its config) and the
enforcer is zero-config, so a mistyped tool name is impossible.
Regenerate config-catalog (timeout-policy -> no-config; tool-web gains
fetch/searchTimeoutMs), the event graph (tools/change loses its
timeout-policy consumer), the ToolDefinition type-equiv block, and a
source-line drift in the cordis services catalog.
The seam's structured-clone boundary admits values JSON does not (BigInt,
Map, circulars), while tool/code-dispatch events must be JSON-appendable —
left unhandled, a sub-call could execute and then fail at logging time.
The bridge now JSON-normalizes binding arguments BEFORE dispatch (a value
that does not survive rejects that one call), so the dispatched form and
the logged form are the same JSON value by construction.
Model-facing tool-call budgets were tangled into each capability's schema
(bash timeoutMs, web_fetch timeout_ms) with no shared home. Add a
tools/execute around-dispatch waterfall to dsh-tools whose base next() is
the dispatch-with-normalization thunk, and a new @deepseek-ai/dsh-timeout-policy
plugin (packages/timeout/) that arms a per-tool deadline on exec.signal and
returns a structured TOOL_TIMEOUT when it wins. Migrate web_fetch (drop the
model-facing timeout_ms) and web_search onto it; the fetch provider keeps its
timeout only as a resource backstop for direct callers. bash and hook command
execution keep BASH_TIMEOUT unchanged.
Named the plugin timeout-policy (not the RFC's tool-timeout) so it does not
trip the gen-tool-catalog packages/*/tool-* completeness guard, and replace
exec.signal by in-place mutation before next() since cordis waterfall next()
ignores passed arguments. RFC moved to implemented/ recording both deviations.
tianyicui's review: the seam name did not say what the event or its
types do. 'advice' reads both ways — advisory content for the model,
and AOP before/after advice woven around a join point (here the
derived history) without modifying it — so RequestAdvice.before/after
are self-describing. Types follow: RequestAdvice / RequestAdviceContext;
the logged EpochHeader fields keep their positional names
(messagePrefix/messageSuffix).
Also sharpens the core.md wording the review flagged as ambiguous:
before-advice sits in front of the ENTIRE derived history, directly
after the system slot (the conventional home for session-stable openers
— an AGENTS.md digest, a skills catalog), after-advice follows the
history's last message. Catalogs and doc graphs regenerated.
A client-callback throw only becomes a JSON-RPC error RESPONSE to the
agent's session/request_permission — runScenario itself kept going, so a
tolerant agent could treat the error as a denial and the scenario would
pass, or worse, record: the impossible click baked into fixture and
golden, green on every replay. The mismatch is now captured as a harness
error while the agent is answered plain cancelled (a well-defined path
it cannot reinterpret), and the step loop rejects the run on it as soon
as the in-flight step settles. The spec asserts the rejection instead of
the agent-side error echo.
Yichen Jiang
Tianyi Cui
imccyu
Dudu-0223
kingwl