1. Cost-reporter SQL: phpcs:ignore annotations were embedded inside two SQL
string literals (get_daily_spend, get_spend_by_stage), causing MariaDB to
reject both queries and break the Metrics & Costs admin page. Moved annotations
above the prepare() calls. Regression tests added to CostReporterTest.
2. Collector failure observability: catch block in collect_from_all_sources() now
logs structured meta (model slug, HTTP status parsed from exception message,
first 200 chars of error body) so outage incidents are diagnosable post-hoc.
3. run_stages.agent_role back-write: Pipeline_Runner, Article_Worker, and
Content_Generator were calling Run_Stage_State::start() with agent_role=''.
All start() call sites now pass Stage_Display_Map::default_agent_role(stage).
Role written at INSERT only; ON DUPLICATE KEY UPDATE does not include
agent_role — idempotency-key semantics on resume are unchanged.
4. Runware float deprecation: normalize_seed() applied % to a float
(microtime(true) * 1000), triggering PHP 8 deprecation on every image
generation. Fixed with (int) round(...) before the modulo operator.
Refs: convo/prautoblogger/threads/2026-06-pipeline-v2-phase1/25-cto-v0182-maintenance.md
Agent-Session: prab-v0182-maintenance-20260612
Pipeline v2 Phase 1, commit 4/4. Run_Stage_State persists per-run
per-stage state keyed run_id + stage + agent_role + item_key (item_key
scopes articles because one run_id spans a whole batch; agent_role is
the Phase-2 quorum dimension, schema-only). Done stages snapshot output
and are reused on re-entry — writer stages via
Content_Generator::set_run_item, the review via its stored verdict, a
published item skipped outright — never re-run, never re-charged. Post
creation is keyed by _prautoblogger_run_id + new _prautoblogger_idea_hash
(check-before-insert in Publisher) so retries cannot duplicate posts.
Run-level research/analysis checkpoints + done/failed lifecycle marks
ride Pipeline_Runner/Executor (incl. orphan-abort + budget-abort paths).
Run_Reaper rides the EXISTING #19 daily cron: stages stuck running >2x
expected stage wall-clock and runs silent >2x expected run wall-clock ->
failed (filterable), open stages reaped, queryable as 'incomplete'.
Retention: request_json + stage output payloads NULLed after R days, R
from the new Audit Payload Retention setting (default constant 14,
0 = keep forever, no literal at point of use). Audit_Writer records
editorial verdicts (run_decisions) and per-article kept sources
(run_sources) for new runs.
Under-300: status-transient/summary helpers extracted verbatim to
Pipeline_Status (Pipeline_Runner 271; removed the worker's duplicate).
Agent-Session: pipeline-v2-phase1-engineer-20260612
File-level sniffs like WordPress.Files.FileName are evaluated at the file level,
not at code level. They require phpcs:ignore directives to be placed within the
file-level docblock (/** ... */), not in code comments.
Moved all 81 phpcs:ignore pragmas from code comments to the file-level docblocks.
This applies to:
- 79 class files with class-naming convention ignores
- prautoblogger.php with strict-types docblock ignore
- uninstall.php with strict-types docblock ignore
- Added phpcs:ignore for class file naming convention (79 violations)
WordPress.Files.FileName.InvalidClassFileName is a known architectural
issue — class files use short names without prautoblogger- prefix
- Added phpcs:ignore for ->prepare() table interpolation (56 violations)
WordPress.DB.PreparedSQL.InterpolatedNotPrepared: dynamic table names
(prab_generation_logs, prab_cost_logs) require interpolation; ->prepare()
does not support table name placeholders, only column values
- Fixed file-level docblock positioning (2 violations)
prautoblogger.php and uninstall.php: moved declare(strict_types=1) before
file docblock per PHP best practices
Run phpcbf --standard=WordPress-Core across includes/, prautoblogger.php,
and uninstall.php. Fixed 1,362 formatting violations in 75 files:
- Short array syntax: [] → array()
- Line breaking for long function calls
- Consistent indentation
- Function call parameter alignment
No logic changes; purely mechanical formatting.
The frontend status poller measured elapsed time from `started` (set once
at the beginning of the run) and declared a stall at 300s. Three articles
take ~6 minutes total, so the poller always false-alarmed on article 3.
Now every stage broadcast writes `last_updated` into the transient, and
the stall check compares idle time since the last update — not since the
run started. This lets N-article runs complete without false timeout as
long as each individual article finishes within 5 minutes.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The pipeline previously ran all articles sequentially in a single PHP
process. On Hostinger, this timed out at ~120s — enough for 1 article
but not 2+. Now the pipeline generates article 1 in-process, then
queues articles 2..N as individual WP-Cron events. Each article gets
its own PHP process with its own execution time budget, so we can
generate any number of articles regardless of server time limits.
Architecture:
- Pipeline_Runner: orchestrate (collect+analyze+score) + first article,
then serialize remaining ideas to wp_options and chain cron events
- Article_Worker (new): focused single-article worker extracted from
the pipeline loop — content gen + editorial + publish
- Executor: new on_process_article_queue() handler for chained events;
status poller re-schedules orphaned queue jobs automatically
- Article_Idea: added to_array() for queue serialization
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
