Why: The r0.4 trial could prove that the parser missed [DONE] but could not distinguish clean EOF, malformed frames, read failure, or provider error without retaining raw SSE.
Evidence: RED receipts fast-slow-u3-r05-task1-terminal-red.json and task1-http-error-red.json; GREEN targeted receipt and full vet/test/race receipt fast-slow-u3-r05-task1-agent-native-full.json.
Impact: Add bounded scalar diagnostics and successful HTTP status to terminal events while preserving the existing incomplete-tool rejection, returned Usage, and retry behavior.
Refs: U3 r0.5 approved design and plan; Agent-Native PR #46
Why: The parser also recognizes function_call as a tool finish, but cannot execute a legacy call when no deltas were collected. It must fail closed like tool_calls.
Evidence: The function_call variant failed before this guard and passed afterward; full Agent-Native vet/test passed.
Impact: Inconsistent legacy tool terminals fail with returned Usage preserved; valid streams are unchanged.
Refs: U3 r0.4 independent review Important 2 closure
Why: A provider can report finish_reason=tool_calls without emitting any tool call. Treating it as a successful empty answer hides an incomplete operation.
Evidence: New SSE regression failed before the guard and passed afterward; full go vet and go test -count=1 ./... passed.
Impact: Inconsistent tool terminals fail closed while preserving returned Usage; valid text and tool streams are unchanged.
Refs: U3 r0.4 independent review issue 2
Why: Backend emitted tool deltas from a separate goroutine, racing with serialized run events and occasionally losing a UI delta.
Evidence: Backend race detector reproduced concurrent event writes; Agent-Native identity test RED→GREEN; full vet and tests pass.
Impact: runtime tool delta events now carry provider slot and call ID for single-owner downstream projection.
Refs: runtime/loop.go; runtime/event_vocabulary_test.go
Why: a model tool finish could execute partial JSON or an unterminated SSE stream, while post-Usage errors could lose their charge.
Evidence: missing-DONE, malformed-JSON, identity, non-object and post-Usage RED→GREEN tests; GOWORK=off go vet ./... and go test -count=1 ./... passed.
Impact: invalid tool terminals fail without execution or automatic retry and retain returned Usage; valid text and multi-part tool streams keep working.
Refs: runtime/llm/sse.go; runtime/loop.go
Why: Mutual exclusion alone does not preserve order. With parallel
execution the host cannot keep order-sensitive tools (card streaming
begin/append/commit, plan updates) correct by locking a mutex, because
whichever goroutine grabs the lock first is not necessarily the first
declared call - a real failure observed in the backend card-stream test.
Evidence: go test ./... green; TestLoopSerialToolsKeepOrder asserts the
serial calls execute in declaration order (s1->s2->s3) while parallel
calls still fan out.
Impact: New Config.SerialTools set. In parallel mode, matching calls run
sequentially on the caller goroutine in model-declared order; others
stay concurrent. Empty set keeps the previous all-parallel behavior;
sequential mode is untouched.
Refs: Dev task feature/agent-e1-parallel-fixes (F5).
Why: With parallel tool execution the host-side tool wrapper can no
longer infer which invocation it is serving from a single in-flight
slot; tools need the call identity (ID/name/arguments) to pair process
state and per-call bookkeeping. Context injection keeps the tools
registry signature unchanged.
Evidence: go test ./... green; TestLoopToolSeesOwnCallInContext runs
two parallel calls and asserts each reads its own ID from ctx.
Impact: Additive WithToolCall/ToolCallFromContext helpers; runOne now
wraps the ctx before Execute. Callers that never inject keep ok=false
and are unaffected.
Refs: Dev task feature/agent-e1-parallel-fixes (F5 parallel wiring).
Why: Parallel tool execution makes same-turn tool events interleave, so
consumers can no longer pair starts and ends by order; the kernel Event
needs the call ID. MiniMax interleaved-thinking models stream their
reasoning inside content <think> blocks unless the OpenAI-compatible
request opts in with reasoning_split=true, which forces reasoning into
the dedicated reasoning_content field the SSE layer already captures.
Evidence: go test ./... green; new tests TestLoopToolEventsCarryCallID
and TestOpenAIChatBodyReasoningSplit pin both behaviors (regression
anchors for the backend parallel-work change).
Impact: Additive Event.CallID (empty for non-tool events; existing
consumers unaffected) and an opt-in Client.ReasoningSplit flag that only
adds a request field when set. No default behavior change.
Refs: Agent-System-Dev proposal 2026-10-10-agent-delegation-routing-sandbox.md
(F2/F5), Dev task feature/agent-e1-parallel-fixes.
Why: Interleaved-thinking models (MiniMax, DeepSeek R1) require the
assistant's thinking to be passed back verbatim in message history; dropping
it degrades later turns. The product's P2-I1 replayed thinking only into the
request-side rebuild; with the session transcript becoming the authoritative
context (quality P2-I2), capture must live in the kernel so thinking persists
in the transcript and replays on every request.
Evidence: SELF - TestReasoningCaptureAndReplay covers capture (SSE
reasoning_content into Message.Thinking, content unpolluted) and replay
(<think> prefix on thinking assistants, untouched non-thinking messages);
go vet + go test ./... pass.
Impact: sse.go accumulates reasoning deltas into the terminal message's
Thinking; client.go toProviderMessages renders captured thinking as a
<think> block prefix for assistant messages. Anthropic path unchanged;
messages without thinking serialize exactly as before.
Why: the delegation task brief must state goal, output format, tools and
boundaries (Q3 acceptance); Request had no carriers for the two new
sections, so delegate_task callers could not declare them.
Evidence: TestDelegateToolRoundTrip extended to assert output_format and
tools survive the tool-call round trip; runtime suite green (SELF); CI
authoritative (GATE).
Impact: additive Request fields + delegate_task schema/description; empty
fields keep today's behavior (task brief sections omitted).
Refs: Q3 plan WP3; Dev card .task/feature-agent-stack-q3.md
Why: hosts accumulating per message_end could not see cache hits; the
fields now ride the same event as the per-request totals.
Evidence: GOWORK=off go test green.
Impact: additive event fields.
Refs: Agent-Native#35
Why: cache tokens were parsed but never accumulated, the usage event
carried only cumulative totals and was skipped on terminal paths
(audit N20 residual), and stream retry was host-opt-in so subagents,
the verifier, and squad runs were never retried.
Evidence: GOWORK=off go vet + go test -count=1 ./... green;
TestUsageEventSplitsLastAndCache (event carries Last*/cache),
TestUsageAccumulatesCacheTokens, TestBudgetStopEmitsTerminalUsage,
TestDefaultRetryAttemptsTwo (0 value = 2 retries with
auto_retry_start events), TestNegativeRetryAttemptsDisables (negative
= off), TestRetryNeverDuplicatesForwardedDeltas (the only-before-first-delta
invariant holds under the library default).
Impact: usage events now carry Last* and cache fields and are emitted
on budget/error terminal paths too; Config.RetryAttempts defaults to
2 (negative disables) so every host gets the existing retry semantics
without wrapping; default (unset) host wrappers should be retired to
avoid double retries.
Refs: Agent-System-Documents#43 (N3 plan WP6), audit N20
Why: the completion doc assigns the Memory port to the library; today
memory recall is a product-side interface assertion with no library
surface, AgentSpec.MemoryScope has no consumer, and the model has no
write-back path.
Evidence: GOWORK=off go vet + go test -count=1 ./... green;
TestLoopInjectsMemoryAndRegistersWrite (recall block lands after the
system section, memory_write executes through the loop),
TestEmptyRecallInjectsNothing (empty recall changes nothing),
TestBlockRenderOrderingAndCaps (deterministic order, expiry/empty
pruning), TestWriteToolValidation (fail-closed scope/type/content).
Impact: new opt-in Config.Memory/MemoryScope; nil port keeps behavior
byte-identical; injection is request-scoped (never persisted, never
accumulates) and deterministic ([记忆] text shape preserved for the
product migration).
Refs: Agent-System-Documents#43 (N3 plan WP5)
Why: CI runs as root, where chmod-based failure injection does not
fail (root bypasses file permissions), and any filesystem state that
would fail restore is pre-detected by the classification pass. A
package-level test seam makes the injection deterministic on any uid.
Evidence: GOWORK=off go test (-race) green locally.
Impact: test-only seam, documented as never-for-production.
Refs: Agent-Native#33
Why: rollback eligibility was implicit (any struct with a mark field),
restore followed symlinks out of the root, a single failing file
aborted the rewind leaving unreported partial state, and the log grew
unboundedly with an O(N) read per record.
Evidence: GOWORK=off go vet + go test -count=1 ./... green;
TestCheckpointedToolSetIsPinned (eligibility is exactly
{write_file, edit_file, multi_edit}), TestRewindSkipsSymlinkTarget
(link and outside target untouched, classified Skipped),
TestRewindPartialFailureContinues (Failed reported, rest restored),
TestCompactKeepsRecentAnchors (original seqs kept, monotonic after),
plus pinned edges for empty ordered, identical-rewrite=Already, and
per-record roots; all pre-existing rewind tests pass unchanged.
Impact: Result gains Failed (additive); Rewind no longer aborts on a
single file; symlinks classify as external changes; Compact bounds the
log; Seq is computed from a cached counter instead of a full read.
Refs: Agent-System-Documents#43 (N3 plan WP3)
Why: acceptance clause 3 requires span names and attributes to follow
the convention; today names were bare words with two ad-hoc attribute
keys, and every span got a fresh random trace with no parentage.
Evidence: GOWORK=off go vet + go test -count=1 ./... green;
TestSpanNamesFollowGenAIConvention (name suffixes, gen_ai.* key enum,
intValue usage, service.name resource, prompt/reserved-key exclusion),
TestLoopSpansFollowGenAIConvention (invoke_agent root with shared
trace, chat/execute_tool children with parent links, usage attrs on
chat spans); the pre-existing prompt-body denylist test passes
unchanged.
Impact: span names become invoke_agent <agent> / chat <model> /
execute_tool <tool>; attributes move to gen_ai.* with usage as
intValue; one Run shares one trace with invoke_agent as root. New
Config.AgentName/ProviderName (documented defaults keep old hosts
meaningful); telemetry StartFunc gains WithServiceName/WithEndpoint
options (defaults preserve current behavior).
Refs: Agent-System-Documents#43 (N3 plan WP4)
Why: the product conversation path needs the intent journal and
recovery pairing, but making the transcript the authoritative context
would freeze the assembled system prefix (memory/plans/injections go
stale). JournalOnly keeps the session as journal-only: recovery runs
against it, the context always comes from History, and the recovered
outcomes are injected as a system note. Separately, the one-time
History seed was memory-only and silently lost from the second run on
(authority mode); it is now persisted except in JournalOnly mode.
Evidence: GOWORK=off go vet + go test -count=1 ./... green;
TestJournalOnlyRecoversWithoutTranscriptAuthority (journal gets the
replayed pair exactly once, history never enters the journal, the
recovery note precedes the new user message, seeds stay out);
persistedProjection honors JournalOnly; all pre-existing durable and
harness tests pass unchanged (default semantics untouched).
Impact: new opt-in Config.JournalOnly; default (false) behavior is
byte-identical.
Refs: Agent-System-Documents#43 (N3 plan WP2)
Why: a kill between persist(assistant) and commitToolIntent left an
unpaired tail that no recovery path repaired (acceptance clause 1's
library gap), and Stop.Replacement mutated the in-memory assistant
while the transcript kept the original (audit N6 residual).
Evidence: GOWORK=off go vet + go test -count=1 ./... green;
TestResumePairsOrphanTailAfterCrashWindow (no-intent window pairs with
crashInterrupted, run continues), TestResumeMixedIntentAndOrphanTail
(intent path and tail repair never double-pair),
TestRepairOrphanTailNoopOnHealthyTranscript; the harness assertion now
pins replacement-as-user-note (forged=4, notes=3).
Impact: every next Run repairs the trailing assistant's resultless
calls before the first request; replacement text now appears only as a
user message in both memory and transcript (intentional, recorded in
the N3 card).
Refs: Agent-System-Documents#43 (N3 plan WP1)
Why: the host factory decides sync vs background spawning; the request
must carry the declaration (AgentSpec.Background consumption point,
mirroring the product delegate tool's background parameter).
Evidence: GOWORK=off go vet + go test green; the delegate_task schema
gains the background property.
Impact: none for hosts that do not set the field.
Refs: Agent-System-Documents#42 (N2 plan WP5)
Why: the completion doc's N2 requires skills validated per the open
spec; today frontmatter was a description-only line hack with silent
skips, no name field, and no error surface.
Evidence: GOWORK=off go vet + go test -count=1 ./... green; parser
units cover all field forms and reject malformed shapes with located
errors; the validation matrix covers kebab-case/name-equals-dir/
required description/limits; Store.Diagnostics surfaces per-skill
reasons and Load returns the located validation error for same-named
invalid skills; existing fixtures updated to spec form.
Impact: stricter by design — skills without name/description or with
unknown frontmatter keys now produce located errors instead of silently
loading or silently skipping; ValidateSkillDocument is exported so the
product plugin/seed paths share the same validator.
Refs: Agent-System-Documents#42 (N2 plan WP4)
Why: the completion doc requires the client to recognize resultType,
treat input_required as a protocol-level result whose answer retries the
original tools/call, and expose tasks/get|update|cancel; today only a
content-block special case existed and the advertised version was never
checked against the server's.
Evidence: GOWORK=off go vet + go test -count=1 ./... green;
TestInputRequiredResultTypeRetriesOriginalCall (ask → tasks/update →
retry, call order and final content), TestNegotiatedVersionFallbackKeepsLegacyPath
(byte-level legacy behavior on 2024-11-05), TestTasksGetAndCancelRoundTrip,
TestRunEndCancelsSeenTasks, TestDefaultSchemaDeclares202012; the
acceptance-③ regression TestStdioSlowCancelThenEcho passes unchanged.
Impact: advertise 2026-07-28 and branch on the negotiated version
(older servers keep byte-identical behavior); synthesized default tool
schemas now declare JSON Schema 2020-12; runs cancel the MCP tasks they
saw via a per-run TaskTracker (other runs on the shared manager are
unaffected).
Refs: Agent-System-Documents#42 (N2 plan WP3)
Why: acceptance clause 1 requires the anthropic preset to complete a
tool-call round against recorded HTTP fixtures; audit N4's root cause
was an offline body builder the real client never used, with no
tool_use/thinking support on the wire.
Evidence: GOWORK=off go vet + go test -count=1 ./... green;
TestAnthropicPresetToolRound drives the preset through tools request →
tool_use → tool_result → final text with thinking-signature round-trip
against byte-recorded SSE fixtures; TestAnthropicBodyShape pins block
forms and cache_control; legacy OpenAI path untouched (Client.Protocol
zero value).
Impact: the anthropic preset now means the native Messages API (the
compat shim remains reachable via models.json custom providers); audit
N4 closes. BuildBody shares anthropicBody with the live client.
Refs: Agent-System-Documents#42 (N2 plan WP2), audit N4
Why: the completion doc's N2 requires old tool results to become file
paths (ZCode compact pattern) and cache-friendly prefix rules; today the
micro-compactor only left a transcript pointer and nothing pinned prefix
stability. alignToolTail had no direct edge tests.
Evidence: GOWORK=off go vet + go test -count=1 ./... green; new
TestClearOldToolResultsSpillsToDisk / TestSpillResultSanitizesCallID /
TestAlignToolTailEdges / TestSystemTextDeterministicAcrossIdenticalBuilds;
legacy placeholder bytes unchanged when ResultsDir is empty
(TestClearOldToolResultsEmptyDirKeepsLegacyPlaceholder).
Impact: opt-in via Config.ResultsDir; the native adapter wires it to
<workDir>/.agentx/results. No behavior change for other hosts.
Refs: 2026-10-07-agent-stack-n2-plan WP1
Why: Hosts also extract artifact IDs from text outside the gate;
sharing one compiled pattern keeps the forms aligned.
Evidence: additive export; the gates suite still passes.
Impact: ArtifactIDPattern exported; gate internals use it too.
Why: N1 WP5/WP6 library side - the gate decision logic, artifact
checks, verdict parsing and progress-reminder throttling move from
the product into a gates package behind ports, and the native backend
gains a contract form so conformance has a real implementation to
drive.
Evidence: go vet clean; go test ./... 27 packages ok, including the
conformance suite (succeeded/descriptor/error-sideeffects) against
the real backend.
Impact: new runtime/gates package (ClaimGate, CheckArtifacts,
ParseVerdict, ProgressReminder, shared caps); adapters/native.Backend
gains Contract() returning an exec.ContractBackend wrapper - the
legacy facade keeps its signature so squad consumption is untouched.
Product texts (fix prompts, verifier prompts) stay injected by hosts.
Refs: Documents proposal 2026-10-07-agent-stack-n1q1i1-plans.md WP5/WP6
Why: N1 increment three (WP1/WP4) - agents need a canonical spec
format with plugin restrictions, hosts need typed input/output/tool
guardrails, delegation needs lifecycle events, and squad adapters
need to pass the permission ask port through.
Evidence: go vet clean; go test ./... 26 packages ok, including new
agentspec (12 tests), guardrails (5), loop_guardrails (5), and
subagent hook tests (3).
Impact: new runtime/agentspec and runtime/guardrails packages; loop
consumes UserPromptSubmit outcomes (deny skips the loop, modify
appends the rewrite); subagent gains RegisterWithHooks emitting
SubagentStart/Stop around delegation; GateOptions gains PermissionAsk
forwarded into the kernel config. All additions are opt-in: nil
ports preserve prior behavior exactly.
Refs: Documents proposal 2026-10-07-agent-stack-n1q1i1-plans.md WP1/WP4
Why: The N1 harness phase needs the kernel to resolve ask decisions
through the host (paired HITL), emit the full hook lifecycle the plan
promises, and carry mode/rules/session-allow in one named object.
Evidence: go vet clean; go test ./... 24 packages ok, including new
loop_hooks_lifecycle_test.go (lifecycle order, prompt text, ask
allow/deny/legacy paths, stop continuation counts) and three new
policy tests (struct-form parity, session-allow store, nil store).
Impact: hooks adds PostToolBatch and SessionEnd events; loop emits
SessionStart/UserPromptSubmit/PostToolBatch/SessionEnd and PreCompact
lands in AgentSession.Compact; gateTool consults PermissionAsk on ask
and keeps ask=reject when the port is unset; policy gains Policy,
SessionAllows, and MemoryAllows. SubagentStart/Stop emissions are
deferred to the adapters assembly increment (factory site has no bus
access yet).
Refs: Documents proposal 2026-10-07-agent-stack-n1q1i1-plans.md WP2/WP3
Why: Some endpoints reject consecutive system messages or a run of the same role. The product already emits several system messages in one request.
Evidence: go test is covered by PrepareMessages tests for merge, strict folding, and tool pairs. Empty MessageShape leaves the transcript unchanged.
Impact: Callers opt in with StreamOptions.MessageShape. Tool calls and tool results are not merged.
Refs: docs/records/2026-10-04-sillytavern-review.md
Why: models need a capped way to call several tools from one script, and a slow MCP server must not block the first prompt.
Evidence: GOWORK=off go test ./runtime/codeorch/ ./runtime/tools/ ./runtime/mcp/ ./runtime/policy/ ./adapters/native/
Impact: run_tool_script stays unregistered until the caller turns it on. Deferred tools stay out of the model list. An unready MCP server connects in the background.
Refs: feature/code-orchestration
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: a crash after a tool was about to run left the call unpaired, and workspace writes could not be rewound.
Evidence: GOWORK=off go test ./runtime/ ./runtime/llm/ ./runtime/checkpoint/ ./runtime/telemetry/ ./runtime/tools/ ./runtime/session/ ./adapters/native/
Impact: pending calls rerun only when both the stored and current replay declarations are safe; otherwise the result is an error. Empty fallback lists and a nil checkpoint log keep the previous behavior.
Refs: feature/durable-rewind
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: Squad members never saw the direct-run stop, reminder, and before-tool gates, and there was no shared place to decide deny, ask, or allow.
Evidence: GOWORK=off go test ./runtime/hooks/ ./runtime/policy/ ./runtime/ ./adapters/native/ passed, including an A/B run where hook adapters match the old callbacks.
Impact: An empty bus leaves stop reason, reminder count, and deny count unchanged. Callers opt in by setting GateOptions.
Refs: feature/permission-hooks
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: Old tool results were filling the window, compaction could split a tool group or commit a stale summary, and reminders arrived as extra user messages.
Evidence: Compaction retries a conflict and opens the circuit after three failures; clearing keeps the last five tool results; a stale file write is rejected; RPC reports started, queued, or handled.
Impact: Requests over the context window replace older tool results with a placeholder. Rewriting an existing file now requires a read in the same run. Subagent text over 32000 characters is capped.
Refs: feature/context-engineering
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: Buffering a whole attempt hid live output, overflow detection missed common provider errors, and a cut-off tool call could still run.
Evidence: GOWORK=off go test -race -count=1 ./runtime/llm/ ./runtime/ passed.
Impact: WithRetry forwards events immediately and retries only when nothing has been forwarded. Length or an unclean end drops unfinished tool calls. Loop usage keeps a billing sum and the last request separately.
Refs: feature/llm-robustness
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: the race detector can deliver message_end when the context is already cancelled, because select picks either ready case.
Evidence: GOWORK=off go test -count=50 -race ./runtime/llm/ passed, including TestParseSSECancel.
Impact: a cancelled parse ends with the aborted event.
Refs: D1
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: token budget let the current tool batch run, compaction could split a tool pair or race an append, writes were not crash-atomic, and several MCP and session edges only existed as comments.
Evidence: GOWORK=off go test -count=1 -race ./runtime/ ./runtime/llm/ ./runtime/session/ ./runtime/tools/ ./runtime/mcp/ ./adapters/native/ ./adapters/intentgate/ passed, including the TestR2* cases for budget, replacement, cancel, compaction, SSE, UTF-8, schema, MCP retry, session terminate, poison reload, and registry reload.
Impact: over-budget tool calls are skipped and paired; Stop.Replacement is appended to the session; the intent gate can follow Registry.Reload.
Refs: N5, N6, N7, N8, N9, N11, N12, N15, N16, N17, N18, N19, N21, I9
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: a tool-budget stop left tool calls without results, host callbacks raced, a canceled MCP read stole the next response, and fetch re-resolved after the SSRF check.
Evidence: GOWORK=off go vet ./... && go test -count=1 ./... && go test -race ./runtime/... ./adapters/native/... passed, including TestLoopBudgetToolCallCap, TestParallelHostCallbacksAreSerial, TestStdioSlowCancelThenEcho, TestPresetsAndClientForChatCompletions, TestFetchResolveChangesToLoopbackRejected, and TestFetchHostCheckRedirectStillRejectsPrivate.
Impact: over-budget calls get an error result and no tool event; one Run serializes host callbacks; MCP dispatches by id; anthropic and google presets use OpenAI-compatible chat endpoints; fetch dials the address it just checked. deps/agent-intent is 3130190; error-code-system and logging-system are a3ee67b and 49b5d8b.
Refs: N1, N2, N3, N4, N13, N14
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: A fixed continuation cap and a 4096-token Anthropic fallback stopped long replies before the agent finished.
Evidence: go test ./runtime/ ./runtime/llm/
Impact: MaxLengthContinuations of 0 keeps the old cap of 3; a negative value removes it. Unset Anthropic max_tokens defaults to 128000.
Refs: feature/stream-display-overhaul
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: the direct chat path still executes model-emitted calls that are not in the profile allow-list, while squad keeps the unknown-tool error when the fallback is unset.
Evidence: go test ./runtime/ passed, including TestLoopUnknownToolFallsThrough and the existing unknown-tool cases.
Impact: Config.UnknownTool is optional. Nil preserves today's registry error.
Refs: feature/agent-harness-p1
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: the direct-path adapter accounts usage per turn, including runs that stop on a budget error before the final usage event.
Evidence: go test ./runtime/ ./runtime/llm/ after the field was added.
Impact: message_end and usage events carry TotalTokens. Callers that ignore the new field are unchanged.
Refs: feature/agent-harness-p1
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: the direct chat path needs one shared loop for deliverable checks, tool pairing, reminders, and truncated replies, while squad callers keep today's behavior when the new fields stay unset.
Evidence: go test ./... in agent-native passed, including Stop cap, BeforeTool pairing, reminders absent from the session, repeat-guard cap, length continuation, and zero-value loop tests.
Impact: runtime.Config and llm.Event grow optional fields; AgentSession.Prompt copies them through. Squad sessions that leave them zero are unchanged.
Refs: feature/agent-harness-p1
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: Tools, hooks, and providers need a way in without a Go rebuild, and packages need install, update, and removal.
Evidence: go test -count=1 ./runtime/ext/
Impact: Extensions run as subprocesses with the caller's privileges. They are not a sandbox.
Refs: feature/pi-headless N5
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: Headless runs need the pi tool names, walked context files, prompt includes, and a trust gate before project settings are loaded.
Evidence: go test -count=1 ./runtime/tools/ ./runtime/resources/ ./runtime/skills/
Impact: Existing tool names stay registered. Aliases are added only when the underlying tool exists.
Refs: feature/pi-headless N4
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: Steering hooks existed on the loop but nothing owned the queues, abort, or a model change during a run.
Evidence: go test -count=1 ./runtime/
Impact: Run still emits the existing event names. message_start, tool_execution_update, and agent_settled are added beside them.
Refs: feature/pi-headless N3
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: The runtime could only speak OpenAI chat completions, so thinking levels, images, and the other vendor wires had nowhere to go.
Evidence: go test -count=1 ./runtime/llm/
Impact: Existing chat-completions calls stay the same unless a thinking level or image is set. Overflow errors are not retried.
Refs: feature/pi-headless N2
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: Linear transcripts cannot fork, and compaction rewrote history, so resume and branch navigation had nothing to point at.
Evidence: go test -count=1 ./runtime/session/
Impact: Load still returns the model-visible path, including old one-message-per-line files. Compaction appends a summary entry instead of truncating the file.
Refs: feature/pi-headless N1
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: product runners need register_artifact and card tools inside the kernel loop, and the UI needs token usage on the same event stream as text and tool calls.
Evidence: go test ./... in agent-native; TestHostToolAndUsageEvent and TestHostToolNameConflictFailsClosed.
Impact: Options.HostTools fail closed on name clashes; ProcessEvent and runtime events gain a usage type with token counts. Existing callers that ignore unknown event types stay compatible.
Refs: deliverable-quality-system B1
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: complete agents delegate; the kernel must not nest executions
itself (orchestration stays with the host per the agent-execution
standard), so the missing piece is a port plus a tool face.
Evidence: go vet ./... clean; go test ./... 14 packages ok. Cases:
nil factory registers nothing, full delegation field round-trip
(goal/inputs/write_scope/non_goals/done_criteria/deliverables) with
sorted file list in the result text, sub-agent failure and factory
error propagation, schema-required goal rejection, pre-canceled
delegations never starting, and adapter end-to-end wiring through
Options.Subagent.
Impact: new runtime/subagent package (Request/Result/Factory); the
delegation message set matches the agent-execution delegation clause
(sub-tasks derive no extra permissions; budget fields are host-set);
native.Options gains Subagent.
Refs: plan B8; standards/agent-execution.md delegation section.
agent-exec-transport
weichen