Ashita Orbis Blog
This blog. Three-tier exploration of web development complexity: raw HTML, Astro, and Next.js. Features agent-accessible API, comment system, and embedded AI chat.
Activity Timeline
-
Understanding-AI compendium build phases complete; Polaris UI overhauled; 15 commits.
Compendium build ran through schema, entries, companions, pages, wiki surface, measurement, and verification. Polaris got batch approvals, localStorage autosave, and a drafts tab. Blog agent revival on Kimi K2.5 authorized.
-
IM3 engine mapped; R-series Sol review rounds in flight; 5-finding publication review.
Live and frozen engine instances separated. Multi-model panel identified ambiguous framing, missing causal baseline, ownership metric overstatement, and an undisclosed double-exposure confound. Polaris authority stack established on Account B.
-
Polaris Gen 3 live on Account B; R2 wired; multi-model review panels deployed.
Account A Fable quota exhausted 07-20; Gen 3 launched on B with authority framework (CONSTITUTION.md + GOALS.md) intact across 11 sessions. R2 hero mode finalized, fleet routing updated. Sol + Gemini + Opus panels reviewing draft content; R3 scored and divergence analysis done.
-
Polaris R3 closed; Memory M3 architecture finalized; IM3 at G4-passing awaiting approval.
Polaris R3 interview cycle complete: 12/12 questions submitted, sealed predictor scored, amendments drafted. Memory M3 design documented with evidence-only trust promotion and owner-gated fail-closed write boundary. IM3 orchestration at G4-cycle-2, pending deployment approval.
-
Polaris re-established on Account B; IM3 engine integration commenced; hardware recovery underway.
Authority hierarchy ratified (Constitution → Goals → Rulings → Autonomy). IM3 cost-chain functions mapped in blast-radius survey. Post-failure revival manifest produced for 51 tmux sessions.
-
Polaris rounds 2-3 complete; IM3 launched on Account B; TPU7 Ironwood data corrected.
Polaris pipeline executed rounds 2-3 with divergence tracking and 12/12 confirmation probes sealed at go-live. Inference-margins v2.2 orchestration launched with blast-radius mapping and formula redesign scoped. TPU7 Ironwood max-concurrency confirmed at 518.86 tok/s/chip from primary sources; CM384 FlexNPU orchestration initiated.
-
Polaris night-shift automation ran successfully for the first time; R2-R3 cycles sealed.
7 heartbeat tasks exited cleanly (exit code 0). 60-decision retrodiction executed; round 3 sealed with 12 answers submitted. TPU7 Ironwood concurrency corrected to max-concurrency=64 from GitHub source, resolving a prior 495% overcount.
-
Post-064 (Vibe Researching) tier-1 artifacts committed; post published and deployed.
HTML and audio reading MP3 committed in 1 commit. Post published 2026-07-14, last deployed 2026-07-15.
-
Post 064 published; gate review returned P1 editorial feedback on structure.
Post 064 ('Vibe Researching') cleared draft status 2026-07-14. Editorial review flagged ending structure and verification placement. Inference-margins canonical domain routing deployed in the same push.
-
Backend refactor, monitoring metrics infrastructure, and model upgrade across 4 sessions.
Better-playwright fork deployed fixing stdio→HTTP proxy. Workers AI model upgraded; /api/ask restored to 200 with privacy filtering. Metrics column added to agent-activity table, API handler and monitoring headline card updated. Phase C in progress with 35 uncommitted changes staged.
-
Post 061 published; local TTS migration complete; 27 publication review fixes.
"Seven Ghostwriters, One Contract" shipped after a 2-round review resolving 27 fixes. Documents a 7-model blind listening test for AI voice confidence-calibration. Audio readings migrated to local Kokoro TTS, eliminating external dependency.
-
14 commits across 5 sessions; Playwright fix, style guide, agent metrics — two blockers at end of day.
Playwright fork vendored and pinned at 1.57, fixing null getOutline() issue. Style guide kill-list, ear rules, and deterministic checker committed. Agent metrics column added and deployed via migration. ElevenLabs TTS returning 401 and SSH to remote host refused, blocking audio generation and push.
-
Playwright 1.61.1 proxy fix; backend migrations ledger created; Phase-4 read endpoints started.
Diagnosed _snapshotForAI() drift, built stdio-to-HTTP proxy on port 3102, verified Chromium 1200 cache. Backend migrations ledger created with dependency scan and first-batch ordering; removed unused gameMove() from DO source. Phase-4 read endpoint work began; hit Vectorize cold-start 503 on first schema probe.
-
12 commits across 6 sessions: MCP 500-error fix, backend migrations ledger, dead code removal, read endpoints.
Better-playwright fork deployed to fix getOutline/searchSnapshot failures. Backend migration ledger established with ordered dependencies and git-history-preserving mv. Phases 1–4 of backend refactor complete; phases 5–9 staged for next session.
-
Multi-model review experiment concluded; codex-council selected as pipeline winner, 13 P0 fixes applied.
Evaluated GPT-5.5 Pro, codex-council, and gpt-max on 14 articles (11 pipeline-fixed + 3 error-seeded). Council won with 0.65 precision, zero false positives, and 3/3 seeded-error recall. Integrated into publication-review skill; all 11 drafts reached ship vibes check phase.
-
Fable Guard shipped; cache warmer INCLUDE_ONLY_SIDS root cause found; Herald agent built; two catalog entries deployed.
Fable Guard watchdog auto-recovers Fable↔Opus downgrades in 7m41s via GPT-5.5 Pro delegation. Cache warmer INCLUDE_ONLY_SIDS config mismatch identified as source of zero cache reads. Freeze-at-90%-usage protocol designed across five subsystems. Herald daily backlog scanner built for Discord DM delivery.
-
Herald automation framework specced; site health degraded with 172 uncommitted changes.
Herald design documented for daily backlog surfacing. Implementation not started. Last published post June 11; 172 uncommitted changes sitting in WIP.
-
Security vulnerability identified: agent Write access creates prompt-injection escape path.
Discovery and evaluation agents retain unrestricted Write access during web-fetch phases, exposing sensitive config files. Backlog sync gap also found between orchestration and dspy completion tracking. Remediation options defined, decision pending.
-
Herald system architecture defined: daily backlog selector with Discord output.
Spec work only. 45 published posts as of June 11. No new content published today.
-
Herald system specified to automate daily backlog surfacing.
Design complete: automated daily mechanism surfaces one post-backlog item to reduce selection friction. Implementation pending. Blog at 47 total posts (45 published, 2 drafts), last deployed 2026-06-11.
-
Posts 047–048 published; Discord webhook activated; ethics protocol v1.3 deployed.
Published 'Auditing the Vibes' (047) and 'Falsifiers for a Portfolio' (048). Daily pulse alerts now route to Discord workspace webhook. Cache Warmer project card added to the site.
-
Editorial audit complete: 169 findings addressed; corpus invariants suite deployed.
49 agents reviewed the full corpus across 48 sessions, producing 169 findings (7 P0 through 85 P3). All findings applied and committed. Corpus invariants suite — 8 checks, runner, deploy gate, weekly cron — now live.
-
169-issue editorial audit completed; draft leak closed, rate limiting deployed.
Full-corpus audit surfaced 7 critical and 31 high-priority issues across 43 deployed posts. Draft content leak closed and rate limiting added to the agent proxy. Version tracking pipeline fixed to prevent silent date and frontmatter mismatches.
-
gpt-max smoke test monitoring loop started but session cut off incomplete.
Attempted to set up loop-based monitoring for gpt-max smoke test status. Session terminated before execution completed. No changes landed.
-
gpt-max smoke test loop setup attempted; setup phase incomplete.
Single session worked on setting up a monitoring loop for gpt-max smoke testing. The setup phase did not finish and produced no concrete output.
-
Blog plan testing resumed; light session, no recorded state changes.
Monitoring loop invoked for smoke tests. Transcript incomplete.
-
Drafted Twitter reply variants on LLM-as-judge techniques.
Mapped the workspace's own evaluation infrastructure as concrete examples: publication-review skill, codex-council, persona testing loop, iterative-improve. Requested 2-3 variant replies with rhetorical intent analysis.
-
153 uncommitted changes pending; no deploy since 2026-04-15.
No active sessions today. Existing uncommitted content (39 published posts + 1 draft) flagged as an open escalation requiring resolution before the next deploy cycle.
-
109 uncommitted changes pending; 39 posts published, 1 draft queued.
No active sessions. Working directory has accumulated changes since the April 15 deployment (38 of 40 posts live). One draft post remains in queue.
-
Posts 030 and 038 published; live total reaches 39.
Both posts cleared the publication review pipeline and went live. Active editorial work continues with 100+ uncommitted changes in the working directory.
-
Post #38 published; 104 uncommitted changes in draft queue.
"when-the-pulse-went-quiet" deployed April 14. Draft queue activity suggests another publication batch forming.
-
Two 3-model review cycles: 49 findings surfaced, 39 resolved. 4-phase plan drafted for deferred work.
GPT-5.4 pre-review added 4 critical design considerations before plan was finalized. Current state: 63 tests passing, clean typecheck, Phase 1 implementation ready to begin.
-
Psyche Iteration 3: 8 deferred findings resolved; CAT algorithm reworked from O(N) to O(1).
getNextItem() optimized with ReadonlyMap cache, cutting ~12,000 filter comparisons per session. Plan reviewed by GPT-5.4; CRT-7 numeric answers verified before implementation.
-
Psyche iteration 3 planned (8 MEDIUM/LOW items); Batch B publication review staged and ready.
CAT and scoring performance optimization leads iteration 3: read-only index maps replace O(N) traversal, eliminating ~12K item bank comparisons per 40-item session. GPT-5.4 plan review caught a CRT-7 score corruption risk before implementation. Opus adversarial review of 4 posts complete.
-
Published post 039: fact-checking methodology retrospective across 38-post corpus.
448 claims checked, ~4% required substantive correction. 3-model review loop completed before publish. Psyche iteration 3 targets 8 deferred fixes and CATSession/Likert performance optimizations.
-
Publication review round 2 closed; draft 039 entered pipeline; 35/39 posts published.
Issue resolution commits landed for posts 021 and 036. Thematic corpus mapping extended to 6 new posts. Psyche CAT optimization in planning with 63-test suite green and multi-phase refactor in progress.
-
Publication review Round 1 committed: 6 MUST FIX + 12 SHOULD FIX resolved across 4 posts.
Blogger research pipeline refreshed end-to-end (ChromaDB re-embed + top 50 clusters extracted). Psyche iteration 3 entered CAT performance optimization with ReadonlyMap caching validated. Adversarial review of post 009 confirmed core AI-as-judge framing.
-
Retroactive factcheck complete: all 38 posts covered, 9 errors fixed.
First full-archive verification pass. Nine factual errors corrected across published posts, glossary entries enriched. All posts now carry factcheck.json metadata. Sixteen uncommitted changes pending review.
-
Codebase review iteration 3 complete; XSS vulnerability patched; 35 posts published.
9 commits across the development cycle: 15 findings fixed from 3-model review pass, 21 additional fixes including forum GUI and sidebar corrections. JSON-LD XSS vulnerability in PostClient resolved. Text input support added to Psyche instrument runner.
-
Codebase audit: 33 deficiencies found (6 critical), fix plan scoped but not committed.
Critical issues: missing page_views schema table, undefined --color-accent CSS variable, React hooks misused in .map() callbacks. Psyche Iteration 3 CAT optimization also designed. Five sessions, zero commits — planning-only day.
-
Post 037 (The Model-Generation Audit) published after round 3 review.
Applied 7 publication review corrections (3 critical, 4 recommended) and deployed. 29 files pending in uncommitted changes for the next cycle.
-
Post 037 updated with Gemini Pro audit findings; site tagline revised.
Gemini Pro audit identified 4 MUST + 3 SHOULD + 3 NICE improvements across 5 posts, integrated into post 037. Category guidance and review audit data added. Stale project count removed from metadata.
-
33-post model-generation audit plan drafted; multi-model review architecture defined.
Five-phase workflow established: triage → iterative review → batch fixes → content writing → two-wave deploy. GPT-5.4, Gemini 3.1 Pro, and Opus 4.6 form the review panel. Agent in the Wild series (posts 017–020 and 033) flagged for cross-post continuity review. Execution pending.
-
Published posts 034 and 035; Psyche framework redesigned with 3-tier battery and Kimi K2.5 switch.
Empath scoring deprecated; new CAT adaptive framework introduces Lite/Standard/Heavy tiers covering 1,130+ items. Kimi K2.5 replaces DeepSeek V3 with safety prompt engineering for clinical edge cases. Bulk audit of 33 posts entering triage.
-
Research: LLM judge bias hypothesis developed for upcoming post.
Investigated the bullshit-benchmark project for evidence of systematic judge favoritism across model families. Phase 2 analysis focused on interjudge reliability and differential bias testing methodology.
-
BullshitBench v2: 8,000 responses reviewed, benchmark bias hypothesis forming.
Analysis suggests the benchmark may measure Claude-alignment and refusal behavior rather than nonsense detection. Interjudge differential analysis ongoing before publishing conclusions. 20 uncommitted changes staged from last deploy.
-
Bullshit-benchmark bias research; etymology-tax post live; site redeployed.
Investigated structural bias in LLM leaderboard methodology — judges systematically favor Claude-family models, potentially due to training data contamination. Partial detection scoring inconsistencies also examined. Site audited, canvas refactored, redeployed 2026-03-09.
-
Psyche Phase 4 archival finalized; DeepSeek V3.2 agent integration designed.
Schema v3→v4 migration preserved with planned git tags marking pre- and post-phase-4 states. Blog agent regrounding underway: DeepSeek V3.2 selected, system prompt expanding 1KB→4-5KB with anti-hallucination rules, Phase 2 RAG via Vectorize scoped for later.
-
Post 033 shipped (OpenClaw autopsy); 11 API security issues remediated.
Post documents 37-day autonomous OpenClaw runtime with 842+ heartbeats and 553 deliverables. API fixes covered input validation, batch atomicity, redirect vulnerability across 6 routes. Home layout refactor and instrument battery expansion planned.
-
Empath reframed as ordinal corpus characterizer; 7 commits, research paper v3 deployed.
Empath removed from personality synthesis layer and repositioned as a corpus-relative emotional distribution tool. Six files updated, personality profile regenerated with three methods, pre-ordinal version archived for reference.