feat(cli): profile ingest runs and split model vs tool time (#249)

* feat(cli): profile ingest runs to find where wall-clock time goes

Add opt-in profiling for `ktx ingest`. Each timed phase, work unit, and
agent loop now records durationMs / step count / token usage in the
trace, and a post-run aggregator rolls them up into a "where did the
time go" report printed to stderr.

Enable per run with KTX_PROFILE_INGEST (1/true -> human table, json ->
raw structured profile) or persistently via `ingest.profile` in
ktx.yaml. The json form emits raw milliseconds, token counts, and a
summary.headline one-line diagnosis so coding agents can parse it
directly; json wins when both env and config request profiling.

- runtime-port: RunLoopMetrics (totalMs, usage, stepCount,
  stepBoundariesMs) plus onMetrics callbacks on text/object generation
- ai-sdk + claude-code runtimes: capture per-loop timing and token usage
- work-unit-executor and stages 3/4: thread metrics into trace events
- ingest-bundle.runner: time worktree / triage / clustering / index /
  reconcile / squash phases and emit the profile in a finally block
  (best-effort; never affects the run outcome)
- ingest-profile: new trace+transcript aggregator with table/json formatters
- config: ingest.profile flag; docs: profiling section in ktx-ingest.mdx

* fix(cli): flush tool-call logs before reading ingest profile

Tool transcripts are appended fire-and-forget so the agent hot path never
blocks on logging. The ingest profiler read them before the writes settled,
so per-work-unit toolMs (and the model-vs-tool split derived from it) could
be incomplete. Track in-flight appends and expose flushToolCallLogs() —
bounded by a timeout so it can never hang — and flush before the profiler
reads the transcript.
This commit is contained in:
Andrey Avtomonov 2026-06-01 15:49:17 +02:00 committed by GitHub
parent 22ddf5524c
commit 21744fc520
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
20 changed files with 1243 additions and 56 deletions

View file

@ -110,6 +110,12 @@ const ingestSchema = z
.prefault({ backend: 'none' })
.describe('Embedding configuration used when ingest adapters need to embed documents.'),
workUnits: workUnitsSchema.prefault({}).describe('Concurrency and failure handling for ingest work units.'),
profile: z
.union([z.boolean(), z.literal('json')])
.default(false)
.describe(
'Print a timing breakdown to stderr at the end of each ingest run. `true` prints a human table; `"json"` prints the raw structured profile for coding agents; `false` disables it. Equivalent to the KTX_PROFILE_INGEST environment variable (`1`/`true`/`json`).',
),
})
.describe('Ingest pipeline configuration: adapters, embeddings, and work-unit policy.');