* fix(cli): survive ktx.yaml version skew and derive repo ownership from disk
Loading ktx.yaml is now tolerant of keys this ktx version does not
recognize: they are stripped from the in-memory config (the file on disk
is never rewritten) and reported by ktx status as non-blocking warnings,
while invalid values on recognized fields still fail hard. Repo
ownership is derived from observed state (a .git directory plus a root
ktx.yaml) instead of a ktx.managed git-config marker, so projects
created by any past or future ktx classify identically. initKtxProject
now runs an explicit foreign-repo pre-check and writes ktx.yaml before
initializing git, so an interrupted init leaves only recoverable
residue instead of a bare .git misread as foreign.
* style(cli): trim comment blocks to constraint-only notes
* docs(agents): require constraint-only code comments
Align the tree with AGENTS.md/CLAUDE.md conventions:
- Rewrite user-facing strings, docs, and tests to lowercase `ktx`
(no bare uppercase `KTX` tokens remain outside literal identifiers).
- Drop the legacy `historicSql` migration path and its now-unused
helpers, per the no-backward-compat rule.
- Remove `as unknown as` / `any` casts: narrow `BaseTool` generics to
`z.ZodObject`, add a typed `createLookerClient`, and delete the dead
`getParametersSchema`/`toAnthropicFormat` pre-AI-SDK helpers.
- Use `InvalidArgumentError` for Commander parse failures.
- Finish the adapter→connector prose conversion in the `ktx.yaml` docs
while keeping the literal `adapters` config key.
* fix(cli): isolate ktx project git repos
* fix(cli): remove inert auto commit config
* test(cli): drop stale auto commit fixtures
* docs: document isolated ktx project repos
* test(cli): keep stale config grep clean
* fix(cli): guide setup away from foreign repos at the project dir
ktx owns the git repo rooted at the project dir and refuses to adopt one it
did not create (the Finding 3 isolation invariant). But setup steered users
straight into that failure: the interactive menu offers "Current directory"
first, and `--no-input --yes --project-dir <repo-root>` created directly in
place — both then threw a generic "Failed to initialize git repository:"
wrapper from deep in GitService.initialize().
Extract the ownership rule into a shared `classifyKtxRepoOwnership(dir)` used by
both GitService.initialize() (the invariant) and the setup wizard (pre-flight
guidance), so the decision derives from one rule. Setup now detects a foreign
repo before constructing GitService and: interactively re-prompts (the user
picks the existing `ktx-project` subfolder), or non-interactively returns a
clean missing-input with the actionable message. The typed foreign-repo error
is also surfaced verbatim instead of being buried under the generic wrapper.
Empty/non-repo current directories still work — only foreign repos are blocked.
* fix(cli): keep classifyKtxRepoOwnership total for non-directory paths
The setup ownership guard runs before the existing not-a-directory check, so
pointing a custom/--project-dir path at a file made classifyKtxRepoOwnership
lstat `<file>/.git`, hit ENOTDIR, and throw — crashing the setup step instead
of returning the friendly "path exists and is not a directory" result.
A path that is a file (or missing) holds no git repo for ktx to avoid, so treat
ENOTDIR like ENOENT and return 'unowned'. The downstream existingFolderState
check still rejects a non-directory with its friendly message, and the
classifier no longer throws raw errno for any caller.
GitService.initialize() used checkIsRepo(), which is true whenever the project
dir sits anywhere inside a git working tree. So when a ktx project lived in a
subdirectory of an enclosing repo, ktx skipped `git init` and silently adopted
the enclosing repo as its store.
Every ktx relative path assumes the project dir IS the working-tree root. During
ingest, wiki/SL pages are written through a session worktree (whose root is the
worktree dir, so the page is recorded at repo-relative `wiki/global/<key>.md`)
and then squash-merged into the main worktree. With an adopted enclosing repo,
the main worktree's root is the enclosing git root, so the merge wrote the page
to `<gitRoot>/wiki/global/` — outside the project dir. reindex scans
`<projectDir>/wiki/global/`, found nothing, and wiki_search silently returned
empty (knowledge_pages = 0) even though ingest reported success.
Detect the project dir's own root with checkIsRepo(IS_REPO_ROOT) and initialize
a dedicated repo there unless the project dir is already a repo root. This keeps
adopting a user-created repo when the project dir IS that repo's root, fixes the
silent wiki/SL/memory divergence at its source for every writer, and stops ktx
from committing its scaffold into the user's enclosing repo.
Regression tests cover both layers: a project nested in an enclosing repo gets
its own .git (and the enclosing repo stays untouched), and a wiki page written
through a session worktree + squash-merge lands in the project dir and is
discovered by reindex.
* feat(cli): add ingest rate limit governor
* feat(cli): wire ingest rate-limit config
* feat(cli): report provider rate-limit signals
* feat(cli): show ingest rate-limit waits
* fix(cli): complete rate-limit event coverage
* fix(cli): abort ingest provider calls cleanly
* fix(cli): propagate ingest cancellation
* fix(cli): reject pre-aborted ingest rate-limit waits
* fix(cli): honor Claude rate-limit reset waits
* fix(cli): retry thrown Codex rate-limit failures
* fix(cli): type Claude rate-limit result details
* fix(cli): emit ingest rate-limit countdowns from rejected signals
* fix(cli): report ai sdk rate-limit header utilization
* fix(cli): gate LLM rate-limit retries on the governor budget
The AI SDK and Codex runtimes retried 429 / opaque rate-limit failures up
to 6-7 times with no backoff when constructed without a RateLimitGovernor
(scan, memory, setup) or with pacing disabled, ignoring Retry-After and
worsening the limit. The outer retry loop only cooperates with the
governor's pause, so without active pacing there is no backoff to apply.
Route the retry bound through a single source: RateLimitGovernor
.maxRetryAttempts(), which returns retry.maxAttempts when enabled and 1
(no outer retry) when absent or disabled. All three runtimes (ai-sdk,
codex, claude-code) now use it, so ingest.rateLimit.retry.maxAttempts
genuinely controls attempts and the hard-coded 6 (plus Codex's off-by-one
extra attempt) is gone. Backend-native retry (e.g. the AI SDK's maxRetries)
still handles transient 429s.
Also correct the ktx.yaml docs for maxWaitMs (caps each wait, not the whole
run) and maxAttempts, and sync uv.lock ktx-sl/ktx-daemon to 0.9.0.
* feat: add codex sdk runner foundation
* feat: parse codex runtime events
* feat: expose codex runtime mcp tools
* feat: add codex llm runtime
* feat: wire codex llm backend
* test: avoid Array.fromAsync in codex runner test
* docs: document codex llm backend
* fix: tighten codex runtime config ownership
* fix: use codex sdk env and thread options
* fix: parse codex sdk event shapes
* test: add codex backend live smoke
* docs: clarify codex backend isolation
* fix: drive codex loop metrics from mcp events
* fix: enforce codex local step budget
* docs: disclose codex isolation limits
* fix: count all codex agent steps and stream step callbacks live
The agent-loop step budget only counted completed mcp_tool_call items, so
built-in command_execution steps (which the public Codex SDK/CLI surface can
still expose) never decremented the budget, letting ingest/reconciliation run
past stepBudget until Codex stopped on its own. onStepFinish was also replayed
only after the whole stream drained, so live work_unit_step / reconciliation
progress appeared stuck until the Codex process exited.
collectEvents is now the single live step accumulator: it counts every
completed agent-action item via a shared isCompletedAgentStep predicate
(command_execution, mcp_tool_call, file_change, web_search), fires onStepFinish
as each step completes, and enforces the budget on that broader count. A
no-tool turn still counts as one step. toolFailures stays MCP-specific, since a
non-zero command exit is normal agent exploration, not a loop failure.
* test: align ingest llm-guard assertions with codex backend
The skip-llm ingest guard message now lists codex as a valid backend and
mentions a Claude Code/Codex session plus a codex setup hint, but this slow
suite test still asserted the pre-codex wording. Update it to match the
production message (already covered by the local-bundle-runtime unit test) and
add the codex setup-line assertion.
* fix: treat codex error:null tool calls as success
The Codex SDK serializes error: null on successful mcp_tool_call items, so
the failure check (item.error !== undefined) flagged every successful tool
call as failed with the empty-payload default "Codex turn failed". This
killed every ingest work unit under the codex backend before it could
produce a patch.
Key on status === 'failed' (authoritative, always set) and only treat a
populated error object as a failure. Add a regression test built from a
verbatim real-SDK event capture.
* fix: default codex backend to gpt-5.5 and report real probe errors
The previous default gpt-5.3-codex is an API-key-only model that the OpenAI
API rejects under ChatGPT-account (subscription) auth, so codex status/setup
failed with a misleading "authentication is not usable" message even though
auth was fine.
- Default codex model is now gpt-5.5 (works on both subscription and API-key
auth); the curated setup picker offers gpt-5.5 / gpt-5.4 / gpt-5.4-mini and
keeps free-form entry for account-specific ids (e.g. gpt-5.3-codex-spark).
- runCodexAuthProbe now distinguishes "model not available" from an auth
failure and surfaces the real API error: collectEvents retains stream
events when the SDK throws on a non-zero exit, and the API error JSON
envelope is unwrapped to its human-readable message.
- The Codex isolation warning now renders inside the clack setup frame.
- Docs updated to gpt-5.5 with a note that *-codex ids require API-key auth.
* fix: require llm.models.default in status and match codex probe remediation
Status reported a project ready when a non-none LLM backend was configured
without llm.models.default, but the runtime (resolveModelSlots) hard-requires
it, so ingest/scan/memory threw after `ktx status` said the project was usable.
buildLlmStatus now fails for any non-none backend missing models.default and no
longer invents a fallback model for claude-code/codex.
Codex probe failures now carry a category-matched fix: a model-access failure
steers the user at llm.models.default instead of the auth/install remediation.
runCodexAuthProbe returns the fix and status consumes it; the message stays
self-sufficient so setup output is unchanged.
Docs: README now lists the codex backend and local Codex auth; ktx-setup.mdx
states --llm-model only accepts codex/default or gpt-*/codex-* ids.
Repaired four doctor fixtures that configured a backend without models.default
(the now-correctly-blocked config) and added coverage for the new behavior.
* 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.
* feat(cli): define full warehouse dialect contract
* test(cli): keep dialect edge tests focused
* fix(cli): stabilize dialect contract foundation
* refactor(connectors): own read-only query preparation
* refactor(connectors): resolve dialects through registry
* refactor(connectors): keep concrete dialect classes internal
* chore(workspace): enforce dialect import boundary
* refactor(cli): resolve relationship dialect at scan boundary
* refactor(cli): use dialect display parsing for entity details
* refactor(cli): use dialect display parsing for warehouse catalog
* refactor(cli): use dialect SQL in relationship workflows
* test(cli): verify solid dialect scan workflow closure
* test: split cli tests from source tree
* refactor(cli): standardize BigQuery scope listing
* feat(sqlite): implement connector scope listing
* test(connectors): cover required table listing
* feat(cli): add warehouse driver registry
* refactor(setup): route scope discovery through driver registry
* refactor(cli): route local query execution through driver registry
* refactor(historic-sql): route dialect support through driver registry
* refactor(cli): test warehouse connections through driver registry
* fix(cli): close driver registry type export gaps
* Improve setup daemon diagnostics
* refactor(setup): centralize rail-prefixed diagnostics + query-history fallback
Extract errorMessage, writePrefixedLines, and flushPrefixedBufferedCommandOutput
into clack.ts so the setup wizard, managed daemons, and embedding/agent steps
share one rail-formatted writer. setup-databases.ts also adds a
"disable query history and retry" option when the schema-context build fails
and query history is the likely culprit, surfaced via a new
failed-query-history-unavailable status.
* fix(cli): carry catalog through the picker so BigQuery/Snowflake/SQL Server scope filters match
The setup picker's KtxTableListEntry was a 2-level { schema, name }, so
qualifiedTableId always wrote db.name into enabled_tables. When BigQuery,
Snowflake, or SQL Server later ran fast ingest, their introspect step filtered
the scope set with scopedTableNames(scope, { catalog: projectId|database, db })
— catalog was non-null on the introspect side but null in the scope refs, so
every entry was rejected, the live-database adapter staged zero table files,
and detect() failed with 'Adapter "live-database" did not recognize fetched
source output'.
Align the picker boundary with the canonical 3-level KtxTableRef:
- Add catalog: string | null to KtxTableListEntry.
- BigQuery/Snowflake/SQL Server listTables populate catalog from the
resolved projectId / database; Postgres/MySQL/ClickHouse/SQLite set null.
- qualifiedTableId emits catalog.schema.name when catalog is non-null
(resolveEnabledTables already accepts the 3-part shape) and
schemasFromEnabledTables now goes through parseDottedTableEntry so it
recovers the schema correctly from both 2-part and 3-part entries.
- Export parseDottedTableEntry from enabled-tables.ts (@internal) for picker
reuse.
Update listTables expectations in all seven connector tests and the setup /
picker test fixtures. Add a picker regression test that covers the
catalog-bearing round-trip (save + refine).
* fix(cli): allow debug telemetry under opt-out env
