Dashboard
Updated 2h ago

Projects

Priority-ranked work queue — 181 completed

Active

23

Blocked

18

Monitoring

1

Completed

181

42 active

Active

23
🔴🖥️
trigger.dev Migration — Full Automation Surface off HatchetHomelab
Queue #7 investing-reviewer REBUILD ✅ 2026-05-30 — the LAST item in the lift-and-shift rebuild queue (#1-7 all DONE). The audit flagged it `partial`: it scored holdings with a Yahoo free-API FORK (`src/lib/shim/yahoo_factor_scoring.ts`) that re-implemented the same 5-factor model the repo already had a canonical typed client for at `integrations/fmp/fmp.ts` — a @rule:single-source-of-truth violation. Step-0 (user 2026-05-29): "use FMP up to the free tier and Yahoo for anything missing." A prior session live-smoked FMP and found two hard blockers that fixed the shape as a HYBRID: (a) full-FMP blows the 250/day free cap (holdings + the 31-ticker Discovery scan ≈ 270-330 calls); (b) FMP's insider endpoint is paid-tier (402 Restricted Endpoint). So: 4 factors (momentum/quality/value/revisions) → canonical FMP (`classifyAndScore` `skipInsider`, ~80 calls/run; real Novy-Marx quality + P/E-vs-5yr-history value replace the Yahoo proxies); insider → Yahoo `netSharePurchaseActivity` overlaid per holding + composite recompute; Discovery scan → Yahoo. Deleted the duplicate Yahoo factor fork; promoted the Yahoo client to `integrations/yahoo-finance/yahoo_finance.ts` (now a shared connector); `discovery.ts` → `src/lib/`. Added `opts.skipInsider` (pure-additive, no behavioral callers) + a graceful per-endpoint Zod boundary to `fmp.ts` (error-object → factor null, never throws). Deployed v20260530.1 (`proj_mmfmuknjvmrmqohoocqy`); live dry-run `run_cmprzthvc...` COMPLETED — FMP scored 11/15 holdings (coverage passed the 50% floor), AI thesis ran; Yahoo 429'd and the degrade contract fired exactly as designed (insider null → 4-factor composites, Discovery empty, 2 warnings, run STILL COMPLETED — FMP is the core). A 2nd dry-run FAILED at the UNCHANGED `RecommendationsSchema` guard because today's Gemini quota pressure forced a weak fallback model that omitted the action `type` — the fail-loud guard working, not a regression. 51 app + 53 fmp tests (skipInsider RED observed) + eval 14/14 (2 fixtures, new BUY-cites-Discovery + insider-null degrade) + tsc + biome + shape + cutover gate 16/16 green. Rebuild mode, no Hatchet cutover. Report: `investing-reviewer-rebuild-report.md`. Burn-in: first scheduled fire 2026-06-01 06:00 PT (confirm the full Yahoo path lands when not rate-limited). NEXT TOP-PENDING = the lift-and-shift rebuild queue is COMPLETE. Remaining migration work: #8 service-archiver (retires apps/backups/), the 3 WSP apps, the ~15 manual apps. Queue empty = "all automations running perfectly in trigger."
🔴🔧
Project: Direct-Hire L3 Recruiting PipelineWe Solve Problems
Phase 1 research shipped 2026-05-05 (4 docs in docs/, ~18.8K words). Ashkaan reviews and approves before Phase 2 starts. Phase 3 architecture pivoted: post + inbound funnel + async assessment (NOT scrape + invite — OnlineJobs.ph ToS bans scraping). Phase 2 (JD + interview loop) builds the Phase 1C-defined assessment as live artifacts. Phase 3 (apps/wsp-recruiter MVP) gates on Phase 2.
🔴🔧
AI Product: Workflow Automation as a ServiceWe Solve Problems
June 11 mentorship session with Sabrina Kay to solidify the business model (Fortune 500 pitch / MannKind AI monetization / de-bottlenecking the venture); review the drafted pre-read email + confirm Xpiri specifics before the call
🔴🔧
Project: MSP Lead Generation CampaignWe Solve Problems
Stand up Instantly outbound per Lead Gen Jay playbook (5 sending domains / 25 mailboxes, 14-21 day warmup, Clay AI qualification before sequencer, 3 entry-point offers tested in parallel); GBP photos + Q&A + booking URL; directory cleanup (BBB, Clutch); send review requests to 3 clients
🟡🤖
Project: Parallel-Shard PrimitiveAI
Dogfood on a real sharded project (likely a trigger-dev-migration slice) and confirm think-step-3.6 heuristic on real scope
🟡🤖
Close + Commit-Gate RobustnessAI
Code merged to main at 2420f6d5e; all 6 test suites green (foundation-report.md). Remaining: confirm /close runs clean on the HAPI surface (CLAUDE_PROJECT_DIR unset) in a real close, and watch the journal corpus to confirm the commit-gate failure class stops recurring. User to mark completed.
🟡🤖
Project: Data Reliability FrameworkAI
Ashkaan walks the Top 30 Stakes Walk (widget-audit.md ## Top 30 Stakes Walk — 3 sessions of 10 rows, ≈30 min each) covering finance / sales / ops / cross-portal sanity. If 0-2 drift, declare portals trustworthy and proceed to team validation. If 3+ drift, expand to the affected source's other rows in the full 365.
🟡🤝
Project: EO Event No-Show Feedback SystemEntrepreneurs' Org
Phase 0 — Meta Business + phone + template approval (user-driven, ~48h Meta review). On approval, build integrations/whatsapp/ + apps/eo-noshow-feedback/ + new Event Feedback tab on Engagement Tracker.
🟡🔧
Project: Claude + MCP Learning Path (Ashkaan first, then WSP techs)We Solve Problems
Complete Anthropic Academy "Claude 101" course and log notes
🟡🔧
Project: User Count (v2)We Solve Problems
Wait for v1 (mid-month-offboarding-billing) to battle-test the KV + override pattern, then resume v2 scope
🟡🔧
Team AI EnablementWe Solve Problems
mcp-autotask-admin DEPLOYED + LIVE at https://mcp-autotask-admin.wesolve.tech (2026-05-29) — custom domain bound + cert, 4 Worker secrets set, verified end-to-end (health 200 / 401 gate / tools-list 10 / live Autotask search_tickets). Connector secret in 1P 'MCP Autotask Admin - Connector Secret'. Remaining (needs your claude.ai browser only): add the custom connector at https://mcp-autotask-admin.wesolve.tech/mcp with the Bearer secret — this is the connector-carrier probe (Bearer header; fall back to /<secret>/mcp path URL if claude.ai won't send a custom header). Optional: connect CF Workers Builds for git-push auto-deploy (worker is live now without it). In parallel: sign up Claude Team (3 admin seats).
🟡🔧
Sales Portal (sales.wesolve.tech)We Solve Problems
Test enrichment with business-domain booking; fix SMTP for lead notifications
🤝
EO LA TalksEntrepreneurs' Org
Rehearse AGD + SERT Block 1 out loud, time it against the 2:30/block budget; confirm EO LA slot length to lock the 3-min vs 9-15-min cut
💰
Asset Protection StrategyFinance
Ashkaan to (a) get written non-coverage determination from Chase Fiscus re Cesar EPLI; (b) engage CPA on Solo 401(k) / cash-balance plan design via Echo Base; (c) raise umbrella + add D&O + raise cyber with broker; (d) tighten NuLegal referral-only positioning (B&P §6155 + Rule 1.5.1); (e) eventually engage CA estate/asset-protection attorney to coordinate with QSBS F-reorg.
💰
Project: NuLegal MarketingFinance
Validate GSC 404 fix (7 pages now 200), fix NAP on existing directories, submit new legal/business directories, monitor GSC weekly, add blog-to-state contextual links; upgrade trial-firm outreach to Lead Gen Jay playbook cadence (4-touch sequence with breakup, dedicated sending domain, Clay-qualified list)
💰
QSBS C-Corp Conversion StrategyFinance
Decide on F reorganization timing; then retain Section-1202-experienced tax attorney to execute the Leto-aware structure
💚
Dental — Bone Loss from Grinding (Second-Opinion Reconciliation)Health
Order toothpaste (imported UK Sensodyne R&P preferred) + custom Michigan splint; chase records from old dentist by 2026-04-30; then book follow-up with new dentist plus periodontist consult with CBCT
💚
Blood Flow Restriction (BFR) Training ResearchHealth
Review writeup, decide on device purchase
🏡
Pool Heating — Extend Swimming SeasonHome
Order Sun2Solar 800 Series 20x40 rectangle 8-mil cover (trim to oval) + Feherguard Surface Rider 24-ft reel (~$400-500 total); decide install path (DIY deck-drilling vs pool contractor)
🏡
Project: Homemade TaffyHome
Test recipes, document iterations, finalize flavor profiles
🎬
Project: Playnite Library Syncmedia
Install extension on gaming PC, create Windmill webhook token, deploy sync_playnite.ts, test end-to-end
🔧
CIS Controls v8.1 Vetting (IG1–IG3) — MNKWe Solve Problems
Send the drafted reply (docs/ig3/2026-05-28_email-12.08-13.09.md) to MNK for 12.08 + 13.09. Then request IG3 evidence per the checklist Evidence Required lists as MNK implements the remaining safeguards.
🔧
IT Department Audit — MannKind CorporationWe Solve Problems
Resume remaining interviews (8 remaining of 16, now including new IT Director Omar Chavez added 2026-05-06); collect outstanding IRL items; analysis & case studies; draft report; tentative target end of May 2026 (extended from Apr 15 — MNK deprioritized audit for CIS IG1+IG2 vetting work)

Blocked

18
🔴🏡
Backyard RenovationHome
Blocked: Jess to get French drain quotes
🔴🔧
Cesar Lezama ClaimWe Solve Problems
Blocked: Marquee's reply to Orchid's $30K counter (sent in response to their $215K demand). Remaining open factual items (CRD complaint status, pre-Oct 2022 meal-break policy, 4 factual discrepancies, Hackler Flynn scope) also require third-party input — nothing actionable solo right now.
🔴🔧
Jon Farzam — Fractional AI Officer EngagementWe Solve Problems
Blocked: Fadi response to revised terms sent 2026-05-06 (12-month initial term, retainer reframed as access to WSP + ongoing discovery, builds as separate Build Sprints, API costs pass-through at vendor list price, default model Claude Sonnet 4.6 + Gemini Flash); or scheduling of 30-min walkthrough call
🔴🔧
Project: RAMO Law — Automated Email Digest SetupWe Solve Problems
Blocked: RAMO Law signing on as client; Elsa granting delegated M365 access
🔴🔧
ProLaw → Clio Migration (RMO engagement)We Solve Problems
Blocked: Clio to assign WSP the first task in the migration workstream; SOW #9218 ($7,500 Pro Implementation, May 4, 2026) needs 12 amendments before Elsa countersigns — see [docs/2026-05-05_clio-sow-9218-review.md](docs/2026-05-05_clio-sow-9218-review.md)
🔴🔧
Operations Maturity AssessmentWe Solve Problems
Blocked: Tyler validates against Evolved training; begin L3 hiring
🔴🔧
Buhler & Associates Client AcquisitionWe Solve Problems
Blocked: Matrix Visual (MVS) signing client agreement; Premier Displays (PDE) decision on proceeding
🟡🤖
Dashboard sync stalenessAI
Blocked: trigger.dev migration (projects/homelab/2026-05-13_trigger-dev-migration/) — the 6 ghBatchCommit-bypass writers re-port to trigger.dev during that migration, which changes the commit path again, so verifying the widget audit + auditing DataHealth thresholds now would be redone post-migration. Open tail resumes when the migration lands: (1) verify a widget-audit run shows all rows match, (2) audit DataHealth thresholds (9-12d staleness rendered instead of `[missing]`), (3) migrate apps/wsp-scorecard/analyze.task.ts.
🟡🤖
Project: Deterministic-over-AI Migration AuditAI
Blocked: hatchet-migration per-script ports applying each script's deterministic-over-AI recommendation in the same Hatchet pass. Resumes when hatchet-migration completes (or when same-session inline migrations surface — those still land directly here).
🟡🖥️
Project: Backup & Automation MonitoringHomelab
Blocked: trigger-dev-migration — backup fleet split across platforms: weekly-backup (#8 service-archiver) + the 5 apps/backup-probes/ probes are still on Hatchet; kv-archiver + repo-archiver already moved. A clean single-platform passive watch resumes once both land.
🤝
Project: EO LA President Bid (2026)Entrepreneurs' Org
Blocked: Andy scheduling the interview and/or sharing more details (date, format, panel composition). Site is live behind Cloudflare Access auth; framing + 22-slide content v2 ready. Resume on any signal from Andy — drafting interview prep doc + dry-run cycles is the next active work.
💰
Project: Personal Will (WillMaker Pro)Finance
Blocked: User to upload first drafts for review
💰
Project: Will Preparation — George HassanzadehFinance
Blocked: Attorney engagement and asset discovery completion
💰
Project: First Residential Rental AcquisitionFinance
Blocked: Deal hunt and financing closure
🔧
Project: Westside Build Correspondence Tool — Pilot ImplementationWe Solve Problems
Blocked: 2026-05-01 — 1.0.1 bundle delivered to Jeremiah. Waiting on his side to rotate the leaked Anthropic + ClickUp keys, integrate the keyring snippet, install 1.0.1, sideload `manifest.xml`, and test first run on his machine. Resume on his reply / first-run feedback.
🔧
WSP Exit Strategy: $15M Ranch GoalWe Solve Problems
Blocked: Bones unresponsive; exploring Anthropic opportunity + solo AI pivot
🔧
Project: Asana → Strety MigrationWe Solve Problems
Blocked: sales-portal; then migrate + clean up Strety dupes
🔧
Project: Ninety.io → Strety MigrationWe Solve Problems
Blocked: Scorecard calibration nearly complete (8 exact, 4 close, 6 need team input); Strety OAuth refresh needed in Windmill

Monitoring

1
🔧
MSP 501 AI Award FilingWe Solve Problems
Window: 2026-07-15