Personal CFO: Weekday Morning Money Briefing Bot
Input read-only balances, transactions, holdings & calendar outflows → run protocol for cash runway, anomaly scan, portfolio drift & fee tally → output one-screen daily report with source attribution
Filter
Experience can combine. Pick one topic, and optionally one subcategory. Models and tools can combine within their layer. A case must match every layer you set.
Fuzzy keyword search
Input read-only balances, transactions, holdings & calendar outflows → run protocol for cash runway, anomaly scan, portfolio drift & fee tally → output one-screen daily report with source attribution
Input Gmail inbox → filter by budget tier and brand fit → output ranked shortlist and log to Google Sheets
Input Gmail negotiation thread + Google Sheets historical deals → agent rebuilds deal state, extracts sponsor intelligence, prices from your own records, builds concession ladder → outputs one-page strategy brief for human send-off
Read the sibling-bot registry → detect stalls against SLA → nag once in shared chat → page you only for human-required issues
Control a pre-registered fleet of bots from a single private Telegram thread, with allowlist boundaries and mandatory confirmation for sensitive actions.
Input: shared /workspace with ops boundaries → Process: weekday cadence inventory, archive, backup, collision flagging → Output: live registry + weekly org review
Input your kitchen inventory with rough expiry dates → rank by what expires soonest → get 3 dinner options with cook times and missing items
Input Gong call records and sales data → read evidence by priority, rebuild timeline, extract buyer quotes → output one-page Notion memo
Input: AI vulnerability RSS feed; Process: scheduled fetch → full-text extraction → GPT parsing → split into items → write to ServiceNow; Output: ServiceNow Security Incident
Input pending topics from Google Sheets → AI generates SEO blog posts → email human approval → revise based on feedback or archive
Input: Amazon Buy Again and last 8 weeks of orders → Process: Filter by repurchase rhythm, price-check, add to cart → Output: Review-ready cart table with price-change flags and substitute options
Input product specs, condition, and max price → AI monitors allowed sites → outputs matching links with notes (no purchases executed)
Input vague bug report with screenshot → separate stated facts from assumptions, record precise environment, reproduce step-by-step on staging with synthetic data → output reproducible pack with evidence or non-reproducible finding
Input: GitHub repo and branch → Process: 7-dimension static security audit → Output: findings.md report with paths, line ranges, and snippets
Input advertiser list → read public ad libraries on Meta/Google/LinkedIn/TikTok → output digest of new, scaling, and retired creatives
Input source slide deck URL and target market parameters → run language/currency/proof/compliance passes → output separate copy with risk register
Sales recording → auto-segment & label → reusable clip index with on-demand lookup
Research one niche listing daily → generate cited Markdown draft from template → human publishes after review
Configure any mail-reading bot as a safety sentinel → scan for injection attack patterns → block high-risk actions and await human confirmation
Input historical content library → Compare 90-day periods to find decayed pieces → Output ranked refresh draft list by recovery per hour
No use cases match these filters.