Commit Graph

  • 03105437b8 Revert "fix(navigation): eliminate VLM hallucination on bottom navigation tabs via structural guard" Marc Mintel 2026-04-29 01:19:18 +02:00
  • b9c29a5a2d fix(navigation): eliminate VLM hallucination on bottom navigation tabs via structural guard Marc Mintel 2026-04-29 01:16:06 +02:00
  • 71310b8e84 fix(perception): resolve UNKNOWN screen classification on explore grid and fix LLM fallback API Marc Mintel 2026-04-29 01:08:29 +02:00
  • 073a90c38c test(e2e): purge remaining lying asserts in home and reels tests Marc Mintel 2026-04-29 01:00:57 +02:00
  • 44fae37cc7 fix(perception): enforce strict candidate filtering for grid items Marc Mintel 2026-04-29 00:55:59 +02:00
  • 48071cc9b8 fix(perception): resolve grid hallucination by using 'tap first post' and prompt tuning Marc Mintel 2026-04-29 00:52:17 +02:00
  • 10a85a91f1 feat(memory): enhance memory learning and application observability Marc Mintel 2026-04-29 00:49:04 +02:00
  • 5bf0053884 refactor(perception): remove all hardcoded structural fast paths Marc Mintel 2026-04-29 00:44:14 +02:00
  • a846462d02 fix(navigation): resolve VLM hallucination on EXPLORE_GRID and optimize GOAP logging Marc Mintel 2026-04-29 00:38:58 +02:00
  • 0dbafd0a82 feat(navigation): implement Anti-Loop Guard to prevent Grand Tour deception Marc Mintel 2026-04-29 00:31:40 +02:00
  • 83e5b94ddf test(unit): fix stochastic flake in autonomous goal weighting Marc Mintel 2026-04-29 00:26:50 +02:00
  • dd8285e1ce test(benchmarks): rewrite benchmark runner and add brain scenarios Marc Mintel 2026-04-29 00:14:29 +02:00
  • ac5d5351a6 fix: eliminate thinking-block poisoning + no-op navigation trap Marc Mintel 2026-04-29 00:06:23 +02:00
  • ad012b4cd4 feat: structural test integrity enforcement — mock ban, brain contract tests, UI change noise threshold Marc Mintel 2026-04-28 23:45:22 +02:00
  • 5fcf1f180b fix: smart extraction of action from verbose LLM thinking output Marc Mintel 2026-04-28 23:35:51 +02:00
  • 9a74d89477 test: harmonize intent strings in verify_success for reels and explore grid Marc Mintel 2026-04-28 23:29:17 +02:00
  • dc4b576bc1 test(e2e): decompose monolithic test suite and fortify semantic guards Marc Mintel 2026-04-28 23:09:04 +02:00
  • e94dfe8c5c test(e2e): purge deceptive pytest.skip masks hiding VLM failures Marc Mintel 2026-04-28 21:49:31 +02:00
  • 7aa6bfccf6 feat: add E2E coverage for GoalExecutor.achieve() — close structural gap #1 Marc Mintel 2026-04-28 21:36:16 +02:00
  • 5fef014cb4 fix: purge 5 remaining E2E lies — dead code, theater tests, ghost skips Marc Mintel 2026-04-28 21:28:42 +02:00
  • 0bdfd999d2 feat(navigation): complete autonomous integration tests and goal weighting Marc Mintel 2026-04-28 19:06:16 +02:00
  • 4ad559e107 feat(autonomy): refactor navigation engine to autonomous goals with TDD Marc Mintel 2026-04-28 18:27:45 +02:00
  • f220e09193 🧪 PURGE: All residual mocks and spies from E2E suite. 100% production-parity enforcement. fix/test-suite-integrity Marc Mintel 2026-04-28 17:53:47 +02:00
  • de2a1c104f fix(navigation): enforce HD Map pre-checks and resolve test inconsistencies Marc Mintel 2026-04-28 13:47:10 +02:00
  • 52c553827f fix(core): add structural sanity guards to prevent post-related VLM hallucinations on search and profile screens Marc Mintel 2026-04-28 10:34:44 +02:00
  • cd64794f55 test(core): enforce 100% TDD parity, eliminate mocks, and harden VLM hallucination guards Marc Mintel 2026-04-28 10:26:11 +02:00
  • bd9148e6e9 fix(tests): purge theater/broken tests, fix Config argparse pollution, fix is_ad() false positive Marc Mintel 2026-04-28 09:36:22 +02:00
  • 1e1bba6b16 fix(perception+brain): story view detection + autonomous prompt fix/dm-engine-config-bypass Marc Mintel 2026-04-27 23:55:09 +02:00
  • 2b992cf2a8 test(RED): expose story view detection gap — ScreenIdentity returns UNKNOWN Marc Mintel 2026-04-27 23:51:04 +02:00
  • c051c3a4c3 fix(dm-engine): 4 safety hardening patches with TDD proof Marc Mintel 2026-04-27 23:43:02 +02:00
  • 3006020106 test(RED): 4 failing tests expose DM engine config bypass & spam bugs Marc Mintel 2026-04-27 23:36:55 +02:00
  • 3b9465a3bc fix(GREEN): semantic match guard kills follow hallucination at 3 layers fix/lying-e2e-follow-verification Marc Mintel 2026-04-27 23:22:00 +02:00
  • 5d50228945 test(RED): expose 5 lying tests in follow verification pipeline Marc Mintel 2026-04-27 23:17:04 +02:00
  • 7277f27fae feat(nav): enforce strict embedding length guards and autonomous brain-first navigation Marc Mintel 2026-04-27 23:09:22 +02:00
  • ee3de811d3 test: add TDD proof that Brain is the primary navigation strategy fix/lying-tests-signature-parity Marc Mintel 2026-04-27 22:55:15 +02:00
  • c93333928a feat: make AI brain the primary driver of all goal-oriented navigation Marc Mintel 2026-04-27 22:51:27 +02:00
  • 12937cb2c1 feat: improve brain prompt to aggressively prioritize scrolling over backing out when trapped Marc Mintel 2026-04-27 22:46:48 +02:00
  • 097a5753f9 test: add live LLM test for brain to prevent hallucination regressions Marc Mintel 2026-04-27 22:44:55 +02:00
  • 9ee6aab831 fix: use correct AI model configuration in brain.py instead of embedding model Marc Mintel 2026-04-27 22:36:17 +02:00
  • da804b174a feat: implement brain-driven dynamic decision making to prevent goap traps Marc Mintel 2026-04-27 22:28:53 +02:00
  • 93175b7caf test: add hallucination benchmark and enforce strict guard for structural targets Marc Mintel 2026-04-27 22:13:52 +02:00
  • e37d92cdfd fix: add structural fast path for following/followers to prevent VLM hallucination Marc Mintel 2026-04-27 22:08:30 +02:00
  • 1c38dabe79 feat: gate DM inbox interaction behind explicit dm_reply toggle Marc Mintel 2026-04-27 21:53:57 +02:00
  • 7b8daa7670 fix: enforce quoted intent for follow to prevent VLM hallucination Marc Mintel 2026-04-27 21:47:40 +02:00
  • a7449a1db3 chore(test): Ruthless deletion of ALL remaining MagicMocks and patches across the entire test suite Marc Mintel 2026-04-27 16:50:26 +02:00
  • 746eeb767d feat(intent_resolver): Vision-First Architecture — Set-of-Mark Visual Discovery Marc Mintel 2026-04-27 15:53:05 +02:00
  • 36a8683643 fix(intent_resolver): humanize content-desc for VLM disambiguation (followers vs following) Marc Mintel 2026-04-27 15:41:29 +02:00
  • 888136f733 test(e2e): prove goap planner breaks infinite routing loops when hd map edges are masked Marc Mintel 2026-04-27 15:31:02 +02:00
  • ae36b6e196 fix(goap): resolve infinite routing loop by feeding masked actions to HD Map pathfinder Marc Mintel 2026-04-27 15:24:10 +02:00
  • e70ce0f52d docs: formalize the 100% LLM Autonomy (Zero Hardcoding) directive Marc Mintel 2026-04-27 15:11:30 +02:00
  • 22ca93c988 refactor(telepathic_engine): ruthless deletion of hardcoded DM and comment edge-case guards to enforce true VLM autonomy Marc Mintel 2026-04-27 15:08:23 +02:00
  • 740f8f1f56 fix(perception): pass device object to intent resolver to activate Visual Critic gate Marc Mintel 2026-04-27 15:05:40 +02:00
  • f148efd2a0 fix(obstacle_guard): prevent softlock in ReelsFeed by scoping feed marker strictness to classic feeds only Marc Mintel 2026-04-27 14:59:47 +02:00
  • ac95dec9d8 feat(perception): implement Vision-Critic validation gate to block LLM hallucinations via cropped screenshot validation Marc Mintel 2026-04-27 14:57:20 +02:00
  • 0b68d4bc77 chore: add debug/ to .gitignore to prevent trace clutter Marc Mintel 2026-04-27 14:52:44 +02:00
  • 8c37290bc3 fix(navigation): tie unread indicator dots to thread container bounds to prevent false positive unread threads Marc Mintel 2026-04-27 14:51:30 +02:00
  • b4bafb59be fix(navigation): enforce strict unread badge detection in structural fast paths Marc Mintel 2026-04-27 14:13:00 +02:00
  • 41450c4eaf fix(navigation): implement zero-trust structural fast paths to eliminate VLM hallucination Marc Mintel 2026-04-27 14:00:14 +02:00
  • e9201e0e30 feat(diagnostics): dump screenshots with xmls and limit retention to 5 Marc Mintel 2026-04-27 11:55:44 +02:00
  • ae046be3b1 perf(perception): bypass heavy VLM verification for memorized high-confidence actions Marc Mintel 2026-04-27 11:50:39 +02:00
  • a2a4a75603 refactor(perception): replace XML length heuristic with VLM screenshot verification Marc Mintel 2026-04-27 11:41:51 +02:00
  • 714c914432 feat(navigation): replace hardcoded button guards with autonomous state-toggle penalty learning Marc Mintel 2026-04-27 11:35:05 +02:00
  • 294403d590 fix(navigation): implement Strict Button Guards to prevent VLM misclassification of user names as follow/like buttons Marc Mintel 2026-04-27 11:19:23 +02:00
  • 117e7a22e7 test(e2e): fix positional arg index in test_llm_false_positive_unlearn due to autospec Marc Mintel 2026-04-27 11:14:21 +02:00
  • 0fbd1b1678 fix(perception): allow state-toggling actions to bypass structural length check Marc Mintel 2026-04-27 11:13:43 +02:00
  • b5cca06ce2 fix: resolve follow.py kwargs and profile obstacle scroll bugs Marc Mintel 2026-04-27 11:13:09 +02:00
  • 3c4dd84a61 chore: add .hypothesis to .gitignore and commit remaining modified files Marc Mintel 2026-04-27 11:01:29 +02:00
  • 9ad49500f9 test(e2e): enforce autospec=True on all remaining patch and patch.object calls Marc Mintel 2026-04-27 10:49:07 +02:00
  • 4de087ae45 test: fix legacy test fixtures breaking plugin evaluations Marc Mintel 2026-04-27 10:19:04 +02:00
  • 42a11107fd test(e2e): eliminate all legacy mocks and establish real-world sim suite Marc Mintel 2026-04-27 01:11:47 +02:00
  • b916b86bc5 fix(e2e): harden test suite — 84 pass, 0 fail, 2 xfail fix/e2e-test-hardening Marc Mintel 2026-04-26 19:25:13 +02:00
  • 0bfda47561 chore: stabilize navigation engine and finalize TDD audit Marc Mintel 2026-04-26 01:43:28 +02:00
  • ddbe8f8e99 fix(perception): Resolve OWN_PROFILE shadowing by OTHER_PROFILE heuristic (TDD) fix-memory-tdd Marc Mintel 2026-04-25 22:35:48 +02:00
  • 5b53a7e4c0 fix(memory): Initialize GoalExecutor singleton with username and validate Qdrant deletes (TDD) Marc Mintel 2026-04-25 22:28:54 +02:00
  • 42eabb7bda fix: implement wipe_all_ai_caches, harden blank_start imports, purge root garbage refactor/plugin-architecture-optimization Marc Mintel 2026-04-25 21:43:53 +02:00
  • 144d6401b5 feat: complete modular plugin refactor with 100% E2E coverage for interactions Marc Mintel 2026-04-25 20:58:07 +02:00
  • 77e8251aa7 fix(sae): stabilize navigation engine, fix container filtering, and negative reinforcement logic Marc Mintel 2026-04-25 13:09:12 +02:00
  • ad1af4edfe fix: navigation death spiral - 3 root causes fix/autonomous-navigation-stabilization Marc Mintel 2026-04-25 11:54:33 +02:00
  • 018b615829 fix: systemic GOAP self-sabotage — SSOT consolidation + step-aware validation Marc Mintel 2026-04-24 22:45:48 +02:00
  • d69da4c974 feat: ScreenTopology HD Map + graph-aware GOAP routing + NAV keyword split Marc Mintel 2026-04-24 22:33:58 +02:00
  • 82bf931b0e fix: VLM guard rejects following-count as 'hallucinated nav tab' + back-press circuit breaker Marc Mintel 2026-04-24 21:58:52 +02:00
  • 8f8efe6f2a fix: eliminate 3 critical nav failures — DM guard, GOAP unlearn, ad escape Marc Mintel 2026-04-24 21:42:27 +02:00
  • 5266b8b290 fix: structural bugs in Grid Fast-Path and Qdrant Memory matching Marc Mintel 2026-04-24 15:56:10 +02:00
  • 87df8d21a9 fix: restore node top-level text/desc extraction for Targeted UX and context correction parity Marc Mintel 2026-04-24 15:48:16 +02:00
  • 6edd2a18fb test: add unit tests for brevity bonus and blank start linguistic match Marc Mintel 2026-04-24 15:20:29 +02:00
  • 2b0d0840a8 fix(physics): disable playful biomechanics during aggressive ad skips Marc Mintel 2026-04-24 14:59:35 +02:00
  • f1590631a1 fix(navigation): stabilize GramPilot autonomous grid navigation and Qdrant persistence Marc Mintel 2026-04-24 13:48:27 +02:00
  • 30724d3c03 chore: FSD stabilization, strict TDD enforcement, and unlearn mechanism Marc Mintel 2026-04-24 13:28:32 +02:00
  • 75009d91a2 feat(navigation): Implement 100% autonomous Blank Start architecture; purge heuristics Marc Mintel 2026-04-22 00:05:03 +02:00
  • fff9ec5b0a Hardening: Streamlined testing infrastructure, unified toolkit, and established English TDD standards Marc Mintel 2026-04-21 10:17:27 +02:00
  • ee3022e95d feat: implement smart unfollow with resonance evaluation and close friends guard Marc Mintel 2026-04-21 02:45:05 +02:00
  • 0a02e901b6 fix(feed): resolve home feed back-button scroll-to-top trap Marc Mintel 2026-04-21 02:00:01 +02:00
  • 2c6404f387 feat: stabilize autonomous instagram bot suite (100% green) Marc Mintel 2026-04-20 15:11:49 +02:00
  • fc3209bdc1 test: stabilize E2E coverage and GOAP fallback logic Marc Mintel 2026-04-20 00:33:27 +02:00
  • ba4d7ffda2 chore: migrate autonomous navigation to GOAP and finalize 100% E2E test stabilization Marc Mintel 2026-04-19 22:14:56 +02:00
  • 0aeed11186 Hardening autonomous navigation: Implemented Modal Guards, Transient Drift Protection (WhatsApp fix), and Aggressive Recovery paths. Cleaned up diagnostic artifacts. Marc Mintel 2026-04-17 12:57:12 +02:00
  • 89f14463c5 update Marc Mintel 2026-04-17 00:44:41 +02:00
  • fa1d01527d init Marc Mintel 2026-04-16 18:58:18 +02:00