Read-Only AI Analyzes Cognitive Exhaust Fumes

Query personal data sources (email, journal, tasks, CRM, browser, notes) with read-only AI to detect cross-source patterns like intention-action gaps and attention drift—safer and more insightful than write-enabled agents.

Cognitive Exhaust Fumes Unlock Cross-Source Insights

Cognitive exhaust fumes are digital byproducts of your thinking—emails, journal entries, tasks, CRM contacts, browser sessions, and notes—that reveal patterns no single tool detects. Analyzing them across six read-only sources exposes intention-action gaps (e.g., planned tasks ignored in browsing), attention drift (e.g., browsing contradicting journal priorities), and relationship blind spots (e.g., unread emails from key contacts). This cross-source synthesis, powered by LLMs like Anthropic's Claude, delivers insights like weekly reflections highlighting commitments, tensions, and omissions, or suggestions for discussing recent readings with network matches based on article topics, CRM profiles, and email history.

To implement, use a GitHub template (https://github.com/shippy/personal-intelligence-kit) with Python scripts that ingest data into structured outputs via API calls, then synthesize in a workspace before exporting to Obsidian, Notion, or text files. For example, a weekly GTD-style reflection script pulls data, prompts for structured summaries (themes, conflicts, notable moments, reflection questions), and generates a Markdown report reviewable in Cursor—taking minutes but providing brutal honesty on thinking patterns, not just productivity metrics.

A cross-source query demo combines browser tabs (via Weaviate SQLite), Clay CRM searches (for AI/European tech/education interests), and email to recommend unread contacts per article, even spotting article authors in your network—all in plain language via Claude skills, consuming high tokens but yielding unique suggestions no isolated tool (email client, task manager, browser) provides.

Read-Only Constraint Beats Agents on Safety and Purity

Write-enabled agents risk unbounded downsides (e.g., nuking relationships via bad emails), while read-only errors cost nothing—you ignore bad analysis. This asymmetry suits high-stakes personal data (career, reputation). Read-only also prevents data contamination: AI writes pollute exhaust with hybrid human-AI patterns, obscuring pure cognition signals. Human-mediated feedback loops preserve agency—you read reflections and act, avoiding AI-drafted responses.

Observers outperform agents per interaction: agents save seconds (e.g., weather checks), but observers reveal weeks of project avoidance. They're distinct categories—a mirror isn't a broken butler—not a stepping stone to agents. Open Claude read-only pales against custom observers for value density, with lower exfiltration and cognitive pollution risks.

Security Risks Demand Examined Trade-offs

Cross-source power creates mosaic effect vulnerabilities: combining fragments paints a full personal picture, making it a high-value hack target. Simon Willison's lethal trifecta persists—private data + untrusted LLM content + external API/shell access enables risks despite no writes. Data sent to Anthropic over open networks exceeds minimal needs. The system isn't fireproof, but deliberate risk assessment (vs. unexamined agent defaults) justifies use. Key lesson: your digital exhaust is your most underused dataset—reflect on it read-only to improve.

Video description
Every other personal AI demo has agents sending emails and managing calendars. I built the opposite: a read-only system that queries my data sources (email, journal, tasks, CRM, browser sessions, notes) but can't modify any of them. This is an intentional limitation. I'll cover why trust asymmetry matters (read is safe, write is dangerous), how cross-source pattern detection beats task automation, and why ""exhaust fume analysis"" of one's cognition is more valuable than yet another AI assistant trying to act on your behalf. Šimon Podhajský - Head of AI, Waypoint AI I'm Head of AI at Waypoint and a full-stack builder with a background in data science and data engineering. I built this personal AI system to scratch my own itch -- and discovered that the ""read-only"" constraint led to better architecture than the agent-first approaches I see everywhere. I made a Github repo with a template for people to try out the read-only AI / personal intelligence system: https://github.com/shippy/personal-intelligence-kit Socials: https://linkedin.com/in/simonpodhajsky https://x.com/sim_pod https://simon.podhajsky.net Slides: https://slides.podhajsky.net/read-only-ai

Summarized by x-ai/grok-4.1-fast via openrouter

5238 input / 1530 output tokens in 10691ms

© 2026 Edge