Commit Graph

17 Commits

Author SHA1 Message Date
iomgaa 4516761dbe feat: record sampling parameters in telemetry (port 20 to 21 fields)
Each of the three emitter entry points has a pinned meaning: only the
attempt path has an effective source, so only it merges extra_body.
2026-07-31 21:30:45 -04:00
iomgaa 32d7869043 fix: harden the observability fields against the verifier findings 2026-07-31 08:28:41 -04:00
iomgaa c2fcd5b1f8 feat: record the observability fields end to end through telemetry 2026-07-31 08:03:43 -04:00
iomgaa 91671a77df test: pin OCR live tests to trust_env=false
httpx reads the macOS system proxy config (not just env vars) when
trust_env is on, and the proxy answers 403 for the LAN service at
10.77.0.20. The raw-httpx case in this file already passed
trust_env=False; the OcrClient cases relied on the default and failed on
any machine with a system proxy enabled.
2026-07-29 23:57:44 -04:00
iomgaa 8559f10403 chore: bump version to 1.0.0 with changelog 2026-07-22 11:18:50 -04:00
iomgaa 3846305634 fix: scope postgres telemetry test to run-prefixed rows
The fixture used to DROP the shared llm_calls table on every run, wiping
concurrent migration-batch telemetry (and its own count assertion was
polluted in return). Assertions now filter by a per-run call_id prefix
and teardown deletes only its own rows.
2026-07-22 10:32:15 -04:00
iomgaa 3c9f306ea3 fix: address M3 verifier findings on soak invariants and probe tests 2026-07-22 02:21:33 -04:00
iomgaa d2138e535f test: add MonkeyOCR live integration suite 2026-07-21 23:08:33 -04:00
iomgaa c7ccb5798c feat: add failure-rate breaker channel with exponential reopen backoff
Dual-channel opening: consecutive failures (CHS-compatible, no streak
bump) plus windowed failure rate (two 30s buckets, min_calls guard).
429s bypass both channels as backpressure, and a probe hitting 429
releases instead of holding the lease. Open duration doubles per
rate/probe reopen up to max_cooldown_s, decaying after stable CLOSED.
record_success gains count_attempt so bad-result successes stay out of
the window.
2026-07-21 09:08:14 -04:00
iomgaa 724dc3c328 docs: backfill env template and migration notes for M2 2026-07-21 01:28:33 -04:00
iomgaa abb65c2324 feat: add postgres telemetry recorder with two-tier degradation 2026-07-21 00:50:33 -04:00
iomgaa 63f2cc294e test: verify cross-connection shared governance state 2026-07-21 00:43:39 -04:00
iomgaa 7ceeab3366 test: add real-wait time-semantics variants for redis backends 2026-07-21 00:43:39 -04:00
iomgaa 0b8460b6d5 fix: address independent verification findings
Classify empty completions as transient per human ruling (fixes flaky
real-gateway smoke and prevents caching empty responses), rename the
factory injection parameter gate to breaker per the frozen design,
rewrite the probe-entry cleanup without except BaseException, declare
python-dotenv explicitly, add a mid-backoff cancellation test, and
record all implementation errata in the design and architecture docs.
2026-07-20 22:01:26 -04:00
iomgaa 893707eb32 test: add integration suite for governance stack and redis cache 2026-07-20 07:53:08 -04:00
iomgaa 4a176b6220 feat: add in-memory limiter and breaker satisfying backend contracts 2026-07-20 06:54:07 -04:00
iomgaa 3058f4c744 chore: bootstrap project scaffolding
Add architecture doc (research-wiki/ARCHITECTURE.md), CLAUDE.md with
tiered SOP for Fable 5, adapted .claude skills/hooks/settings, package
skeleton (src/polygateway), pyproject with import-linter contracts,
Makefile, .env.example and smoke test.
2026-07-20 00:49:10 -04:00