Why: vet of this module failed once intent started importing NFKC, because the sum was only in the intent module.
Evidence: GOWORK=off go vet ./... and go test -count=1 ./runtime/... ./adapters/... ./exec/... passed after go get golang.org/x/text@v0.29.0.
Impact: go.mod gains an indirect require and go.sum is created. Runtime behavior is unchanged.
Refs: I4
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: native must build against the intent head that deep-copies specs and hides private catalog metadata.
Evidence: deps/agent-intent is 05e74d8, the feature/agent-stack-r2 head whose race tests passed.
Impact: the nested intent gitlink moves; native behavior otherwise stays on 17f2a4e.
Refs: I4, I7, I8, I9
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>
weichenandCursor