[2a86d1f5] CI-watch: fix the CI regression on roboco-api (#563)

* [e530aa5e] Diagnose and fix roboco-api CI failure (run 29629255153) (#561) (#562)

* [e530aa5e] fix(tests): narrow None before indexing validate_init_data() result in telegram_initdata self-check

CI run 29629255153 failed on mypy, not the historical pydantic-settings
issue (uv.lock already pins 2.14.2). The __main__ self-check block in
test_telegram_initdata.py indexed the dict[str, object] | None return
of validate_init_data() without narrowing away None first.

* [e530aa5e] docs(qa): document CI fix for mypy type narrowing in telegram_initdata test

Explains the root cause (mypy type error in __main__ block), the solution (None narrowing before indexing), and the safe pattern for future test self-checks that call functions returning optional types.

---------

Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech>
Co-authored-by: Backend Documenter <be-doc@roboco.tech>

* [0884b737] Diagnose and fix Python quality gate + e2e lifecycle smoke CI failures on PR #563 (#564) (#565)

* [0884b737] fix(tests): isolate ROBOCO_SDK_URL for scripted e2e-smoke agents

tests/e2e_smoke/harness.py already isolates ROBOCO_AGENT_TOKEN from the
host environment (the #503/#504 fix) but left ROBOCO_SDK_URL leaking
through. flow_server/do_server both default it to
http://localhost:9000 and forward every rejection there for the
per-verb circuit breaker; inside a real spawned agent container that
port is a live SDK loopback, so the breaker records genuine attempts
for the ephemeral test-agent IDs and trips circuit_open mid-test
(test_sandbox_on_demand.py::test_request_sandbox_guard_chain_over_real_api,
which deliberately causes 3 rejections in a row). Point it at a
guaranteed-refused loopback address so every environment gets the same
fail-open bypass a bare CI runner already gets by having nothing
listening on 9000 at all.

* [0884b737] docs(changelog): document e2e-smoke harness ROBOCO_SDK_URL isolation fix

Document the fix that isolates ROBOCO_SDK_URL in the ScriptedAgent harness to prevent the per-verb circuit breaker from leaking state into ephemeral test-agent identities when the e2e-smoke suite runs inside a live agent container. This ensures the suite passes consistently regardless of whether it runs on bare CI or inside a spawned agent.

---------

Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech>
Co-authored-by: Backend Documenter <be-doc@roboco.tech>

* [3b9a1771] Diagnose and fix ALL make quality + e2e-smoke stage failures on PR #563; confirm real CI green (round 3) (#566) (#567)

* [3b9a1771] fix(e2e-smoke): match real embedding dimension when seeding fake journal chunk

test_c3_deleted_journal_unindexed inserted a 4-dim placeholder vector
into chunks_journals, but the e2e stack's app lifespan eagerly creates
that table with the real settings.embedding_dimensions (1024) before
the test runs, so the insert failed with "expected 1024 dimensions,
not 4". Derive _SMOKE_DIM from settings.embedding_dimensions instead
of a hardcoded constant so the seeded vector always matches the
table's actual column width.

* [3b9a1771] docs(qa): document e2e-smoke embedding dimension fix in round 3 CI diagnosis

Recorded the root cause, solution, and pattern for the final e2e-smoke test failure found in comprehensive sandbox testing: the test seeded a 4-dim placeholder vector but the app's eager lifespan init created chunks_journals with the real 1024-dim embedding column. Updated _SMOKE_DIM to derive from settings.embedding_dimensions instead of a hardcoded constant.

---------

Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech>
Co-authored-by: Backend Documenter <be-doc@roboco.tech>

---------

Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech>
Co-authored-by: Backend Documenter <be-doc@roboco.tech>

docs/backend/qa/ci-e2e-smoke-fix-embedding-dimension.md

@@ -0,0 +1,55 @@
+# CI Fix: E2E Smoke Test Embedding Dimension Mismatch
+
+**Round 3 — Resolved**
+
+## Root Cause
+
+The e2e smoke test `test_data_integrity.py::test_c3_deleted_journal_unindexed` was seeding a 4-dimensional placeholder vector into `chunks_journals`, but the e2e stack's app lifespan eagerly initializes every OptimalService plugin (including JOURNALS) at startup, which creates that table with the **real configured embedding dimension** (1024, from `settings.embedding_dimensions`) before the test ever runs. The insert failed with:
+
+```
+asyncpg.exceptions.DataError: expected 1024 dimensions, not 4
+```
+
+The test's own comment ("the table is created fresh per run") was stale: the app doesn't create it fresh on each test, it eagerly creates it once at startup.
+
+## Solution Applied
+
+Changed `_SMOKE_DIM` from a hardcoded constant to derive from the real runtime configuration:
+
+```python
+# Before
+_SMOKE_DIM = 4  # tiny embedding dim — the table is created fresh per run
+
+# After
+_SMOKE_DIM = settings.embedding_dimensions
+```
+
+Now the seeded placeholder vector always matches the table's actual column width, regardless of the configured embedding dimension.
+
+## Impact
+
+- **Scope:** Test infrastructure only (e2e smoke test)
+- **Risk:** Minimal — single constant reference change in the test module
+- **Behavior:** No change to the code under test; the e2e module itself is unaffected
+- **Verification:** Provisioned real postgres+redis sandbox matching `.github/workflows/e2e-smoke.yml`, ran full `make e2e-smoke` suite (51 tests), all passed
+
+## Pattern
+
+For future e2e smoke tests that seed placeholder data into a schema initialized at app startup, source placeholder dimensions from the runtime settings, not hardcoded constants:
+
+```python
+from roboco.config import settings
+
+# Correct: placeholder always matches runtime schema
+placeholder_dim = settings.embedding_dimensions
+```
+
+Do not assume the schema is created fresh per test — eager initialization hooks may create it once at app startup, before any individual test runs.
+
+## Session Summary
+
+Round 3 comprehensively diagnosed every failing CI stage:
+- All 14 `make quality` stages were already clean
+- The sole real failure was this e2e-smoke embedding-dimension bug
+- All code-level failures from rounds 1–2 (mypy regression, SDK URL isolation) were already fixed
+- This final fix clears both the Python quality gate and e2e lifecycle smoke checks for PR #566

docs/backend/qa/ci-mypy-fix-telegram-initdata.md

@@ -0,0 +1,28 @@
+# CI Fix: Mypy Type Narrowing in Telegram InitData Test
+
+**Run 29629255153 — Resolved**
+
+## Root Cause
+CI run 29629255153 failed on the `mypy` quality gate, not the historical pydantic-settings issue. The test file's `__main__` self-check block at `tests/unit/utils/test_telegram_initdata.py:122` was indexing the return value of `validate_init_data()` (which returns `dict[str, object] | None`) without first narrowing away the `None` type.
+
+## Solution Applied
+Narrowed the `None` possibility before indexing:
+- Assign the result to a variable
+- Assert it is not `None`
+- Then index the dict
+
+This mirrors the existing safe-indexing pattern at line 41 of the same test file.
+
+## Impact
+- **Scope:** Test infrastructure only (no production code modified)
+- **Risk:** Minimal — single-line change in a self-check block
+- **Behavior:** No change to the module under test
+- **Verification:** Local `make quality` run confirmed full pass (mypy clean, 13,383 tests passed, 94.46% coverage)
+
+## Pattern
+For future test self-check blocks that call functions returning `T | None`, always narrow before indexing:
+```python
+result = some_function_returning_optional()
+assert result is not None
+# now safe to access result["key"]
+```