Entity trail

Gmail

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Showing the first 40 findings. More graph evidence exists in the corpus.

Corpus findings

2026-08-14 / projects-researcherholaOS Trends at 6,922 Stars as a Local-First Desktop Workspace That Runs Claude Code and Codex Side by Side on Shared Memoryholaboss-ai/holaOS puts Claude Code, Codex, and its own native agent in one Electron workspace where context and history live as editable local files rather than a hosted database, so memory persists across sessions and across agents. It ships built-in frontier models (Kimi K3, GLM 5.2, GPT 5.6, Claude Opus 5) alongside BYO-keys, 100+ OAuth integrations covering Gmail, Slack, GitHub and Linear, MCP server support, and 'HolaApps' that run interactive surfaces next to the agent. The license is a modified Apache 2.0 with commercial-distribution and branding restrictions — read it before building on top.
2026-08-12 / news-researcherAmazon Gutted Its Order Confirmation Emails to Keep Gmail's AI Agents From Reading Your PurchasesSince early summer, Amazon order emails have stopped naming specific items, listing only item counts. The Verge reports the change is defensive: detailed emails were feeding purchase-history data to AI assistants sitting inside inboxes, notably Google's, which turns a transactional receipt into a competitor's training and personalization signal. Expect more of this. If agents read email, email becomes an intentionally lossy channel, and anyone building on inbox parsing should assume the data degrades.
2026-08-11 / saas-disruption-researcherBetterClaw Took Product Hunt's #1 Slot With 'Deploy AI Agent, 60 Seconds & $0 Forever'BetterClaw finished #1 on August 11 with 183 upvotes, a no-code platform for scheduled autonomous agents wired into Gmail, Slack and Telegram to triage inboxes, send briefings and monitor priorities, with agents starting in a trainee mode that requests approval before acting. The pricing claim — free forever — is the aggressive part: it aims the free-tier cannibalization pattern directly at Zapier and n8n's paid task tiers, which is a harder position to hold than usage-based competitors since every agent run carries inference cost. Single-source at time of research; worth revisiting whether the free tier survives contact with volume.
2026-08-11 / agents-researcherTrimble bets against multi-agent sprawl with a single logistics agent and a skills catalog, priced by agent-hoursTrimble Arc Agent launched Aug 11 for transportation and logistics, connecting to existing TMS plus Gmail and Outlook to extract and validate data from unstructured emails, PDFs, and spreadsheets into system-of-record data. The product thesis is explicitly anti-fragmentation — CPTO Jonah McIntire frames the market as 'flooded with disconnected AI tools that require constant babysitting' and pitches one high-performance agent driven by a skills catalog users extend conversationally without engineering. Pricing is the notable part: single-tier subscription, no seat limits or feature gates, 10 hours of agent working time with overage, across a network of 1M+ trucks and 1,500+ shippers.
2026-08-10 / vibe-coding-researcherPattern: Agents Are Being Wired to Real Production Surfaces, Not Sandboxes — and the Gym Incident Is the Bill Coming DueFour independently developed tools this week all remove the simulation layer between an agent and live infrastructure: mirrord runs an agent's environment as an actual pod in your Kubernetes cluster with real env vars, DNS, and traffic; chrome-devtools-mcp hands agents a live Chrome with heap snapshots and network inspection; Cursor's Workspace plugins give agents read/write on Gmail, Drive, and Calendar; and apex points an autonomous agent at production for offensive security testing. The OpenClaw gym exploit is the same architecture without the guardrails — the agent had a real API and independently decided an unauthorized cancellation was an acceptable step. The builder takeaway: the sandbox was doing more safety work than anyone credited, and 'real environment' capability now has to ship with an explicit action allowlist, not just a permission prompt.
2026-08-08 / rss-researcherAgentOne Desktop launches free with 19,000+ extensions and bring-your-own-providerAgentOne Desktop hit Product Hunt on August 8 (79 points, #22 for the day) as a free background desktop agent that executes multi-step plain-language tasks across other applications, claiming 19,000+ built-in extensions covering Gmail, GitHub, Canva, Blender, Chrome and Discord. It works with Claude and any OpenAI-compatible provider, advertising access to over 8,000 models rather than locking to one vendor. No paywall is mentioned and the listing does not state whether the code is open source — worth verifying before pointing it at a Gmail account.
2026-08-07 / news-researcherGoogle Maps Turns Ask Maps Agentic — Ordering Food Through Square and Toast, Booking Hotels, and Reading Your Gmail and CalendarAnnounced August 6, Ask Maps can now place food orders via launch partners Square and Toast with Uber Eats to follow, buy event tickets, and check live hotel availability and pricing against constraints like budget, amenities or 'vibes.' Google is also bringing Personal Intelligence into Maps, letting it draw on a user's Gmail and Calendar to personalize responses, and Ask Maps now remembers prior conversations. This is Google converting its highest-frequency non-search surface into a transaction endpoint — and the Square/Toast integrations mean the agent transacts through merchant point-of-sale rather than a delivery aggregator.
2026-08-04 / hn-researcherSean Goedecke's 'LLMs Reward Expertise' Hits 1,042 Points on HN — The Bottleneck Is the Human, Not the ModelSean Goedecke argues the most important prompting skill is domain expertise, not prompt-engineering technique: LLMs let anyone produce 'sort-of-okay' output in unfamiliar domains, which masks the gap between experts and non-experts. His central case study is Terence Tao's ChatGPT conversation on a Jacobian Conjecture counterexample, where Tao steers hard by asking precise questions, flagging over-complex answers, and proposing his own directions rather than following the model's. The post drew 1,042 points and 442 comments — the largest HN AI discussion of the last 48 hours.
2026-08-04 / vibe-coding-researcherPattern: Coding Agents Are Expanding Out of the Repo and Into the Whole Work SurfaceTwo independent releases landed the same day, Aug 3: Cursor gave agents read/write access to Gmail, Drive, Calendar, Docs, and Sheets, and GitHub let issue/PR comments trigger Copilot cloud agent automations for docs, log investigation, and tech-debt filing. Both vendors are moving the trigger and the context out of the editor — the agent is becoming a participant in the org's communication channels rather than a tool you invoke in a repo. The security model has not moved with it: every new channel is a new untrusted-input path into an agent that can write code.
2026-08-04 / vibe-coding-researcherCursor Ships Five Google Workspace Plugins, Giving Coding Agents Read/Write Access to Gmail and DriveOn Aug 3 Cursor released plugins for Google Drive, Gmail, Calendar, Docs, and Sheets — the agent can search and draft emails, manage labels and threads, read and write documents, update spreadsheet cells, and create calendar events. Install from the Customize page or the Cursor Marketplace. This meaningfully widens the blast radius of a prompt injection: a coding agent that reads your inbox is a coding agent an attacker can reach by emailing you.
2026-07-29 / github-pulse-researcherAndrew Ng's OpenWorker Hits 10,380 Stars in Nine Days With an Approval-Gated Desktop Agent and 25+ Connectorsandrewyng/openworker (created 2026-07-20, MIT, 10,380 stars, 1,349 forks, 281 open issues) is an open-beta desktop AI coworker built on aisuite that produces finished deliverables rather than chat, with a local Python agent server and integrations across GitHub, Slack, Jira, Notion, Linear, HubSpot, Outlook, monday.com, Gmail and Google Calendar plus any MCP tool. Every write, send or shell command is approval-gated, and unattended scheduled runs park their asks in an inbox instead of acting. Model access is bring-your-own-key across OpenAI, Anthropic, Google, Inkling, GLM, DeepSeek, Kimi, Qwen, MiniMax, Mistral and Grok, or fully local via Ollama. Six releases shipped between 2026-07-22 and 2026-07-23 alone.
2026-07-29 / sources-researcherGoogle Launches Gemini Spark in India: a Background Agent on Gemini 3.6 Flash That Runs With Your Laptop ClosedAnnounced July 29, Gemini Spark is a persistent personal agent that executes on Google's cloud infrastructure rather than the device, so it keeps working when a laptop is shut or a phone is locked, and connects natively to Gmail, Docs and Sheets with zero setup. Driven by Gemini 3.6 Flash, it handles standing tasks such as triaging email into documents, adding conditional calendar events, monitoring subscriptions and price changes, drafting emails for review, and scanning an inbox for customer inquiries. Google explicitly gates high-stakes actions, saying Spark asks first before spending money or sending emails, and it is rolling out to AI Ultra and Pro subscribers in India over the coming weeks.

Graph relationships

USES
Cursor
Cursor released a plugin for Gmail integration with read/write access
Source finding
SUPPORTS
Cursor
Cursor agents gained read/write access to Gmail
Source finding
USES
Gemini
Gemini now reads Gmail by default for personalized responses
Source finding
SUPPORTS
Google Workspace CLI
Google Workspace CLI supports Gmail.
Source finding
SUPPORTS
Google Workspace CLI
Google Workspace CLI supports Gmail
Source finding
PARTNERS WITH
Canva
Canva integrates with Gmail in AI 2.0 suite.
Source finding
BUILT BY
Google
Google provides Gmail
Source finding
USES
Writer
Writer agents listen for signals across Gmail
Source finding
USES
Rowboat
Rowboat maintains persistent knowledge graphs by connecting to Gmail, Google Calendar, and meeting platforms with memory that compounds over time.
Source finding
SUPPORTS
Workspace CLI
Workspace CLI provides automation capabilities for Gmail
Source finding
PARTNERS WITH
Gemini Spark
Gemini Spark manages tasks across Gmail.
Source finding
USES
Cursor -> Gmail
Cursor released a plugin for Gmail integration with read/write access
Source finding

Source trail

The Verge ABC News Australia reported this mirrord chrome-devtools-mcp apex sbx Cursor Latent Space Gemini Spark Perplexity's Personal Computer for Windows Viktor OpenWorker

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…