# AI use cases for Ambient Context (dragthelake/ambient-context) Repo: [dragthelake/ambient-context](https://github.com/dragthelake/ambient-context) | Use cases: 5 | Page: https://dedupedai.com/use-cases/dragthelake-ambient-context/ Ambient Context is a macOS menu bar app that records text from your focused window into plain markdown files so AI assistants can read what you worked on. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Give your AI assistant the day's work without pasting it: https://dedupedai.com/use-cases/dragthelake-ambient-context/brief-ai-assistant-on-the-day/ - Draft a standup from what you actually did: https://dedupedai.com/use-cases/dragthelake-ambient-context/draft-standup-from-what-you-did/ - Find the page or file you saw earlier today: https://dedupedai.com/use-cases/dragthelake-ambient-context/find-page-or-file-seen-earlier/ - Build a cited knowledge base of people and commitments: https://dedupedai.com/use-cases/dragthelake-ambient-context/build-cited-knowledge-base-of-people/ - Tighten capture rules after reviewing a day: https://dedupedai.com/use-cases/dragthelake-ambient-context/tighten-capture-rules-after-review/ ## How to give your AI assistant the day's work without pasting it with Ambient Context Repo: [dragthelake/ambient-context](https://github.com/dragthelake/ambient-context) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Register the app as an MCP server so your assistant can read the day's record and notes directly. **Who it is for:** A knowledge worker who opens an AI assistant in the afternoon and has to re-explain the emails, tickets and calls from the morning. **The problem:** The context from the day sits on screen and never gets written down, so the assistant starts with a fraction of what you saw. **The outcome:** Your assistant reads the day's record and the Notes page through MCP, so it can answer from what you actually worked on. ### Steps 1. Download the signed, notarised DMG from Releases, drag Ambient Context.app to Applications and open it. 2. Grant Accessibility when asked, then choose a folder to save the record in. 3. Run `claude mcp add --scope user --transport stdio ambient-context -- "/Applications/Ambient Context.app/Contents/MacOS/ambient-context" mcp` 4. Open the Agent tab, pick an agent CLI and press Process day on a recorded day. 5. Ask your assistant what you worked on today and check that its answer cites the record. ### Prompt to paste into your agent ```text Using the ambient-context MCP server from the dedupedai repo of dragthelake/ambient-context, read the day with the reading-days tools and summarise the work in five bullet points. Expected output: five bullets, each tied to a block in the record. Verify that every bullet names an app or file path that appears in the Days folder for that date. ``` ### Caveat Read tools work whether or not the app is running, but capture only happens while the eye in the menu bar is open. Risk, credentials: Credential-shaped strings are scrubbed only by heuristics, so check the folder for leftover secrets before an agent reads it. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/dragthelake-ambient-context/brief-ai-assistant-on-the-day/ ## How to draft a standup from what you actually did with Ambient Context Repo: [dragthelake/ambient-context](https://github.com/dragthelake/ambient-context) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Turn the day's written record into a standup update grounded in what was on screen. **Who it is for:** An engineer or team member who has to report yesterday's work at standup and cannot remember which tabs and threads they touched. **The problem:** Recalling a full day of reading, replies and decisions from memory is slow and leaves out the threads you only skimmed. **The outcome:** A standup draft built from the day's Notes and Summaries page, with citations back to the record. ### Steps 1. Record during the day with the eye open in the menu bar. 2. Open the Agent tab and set the time you want each day processed, or open the day and press Process day. 3. Go to Settings > Agent skills and install the bundled skills. 4. Ask your agent to use the standup skill for the date you need. 5. Read the draft against the cited blocks in the record before you speak it. ### Prompt to paste into your agent ```text In the dragthelake/ambient-context repo, use the standup skill to draft a standup for from the Summaries file for that day. Expected output: three short sections titled Yesterday, Today and Blockers. Verify that each item traces to a block in Days// before it is spoken. ``` ### Caveat The standup is only as complete as the record, since the app skips secure password fields and some apps expose little text. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/dragthelake-ambient-context/draft-standup-from-what-you-did/ ## How to find the page or file you saw earlier today with Ambient Context Repo: [dragthelake/ambient-context](https://github.com/dragthelake/ambient-context) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Search the day's record for an excerpt and open the document path or URL it points to. **Who it is for:** A person who remembers reading something earlier in the day but cannot recall which app, tab or file it was in. **The problem:** Screen content is gone once you switch away, so finding it again means retracing the whole day by hand. **The outcome:** The matching block in the record, with the file path or URL where the app exposes one, so you can open the original. ### Steps 1. Record during the day with the eye open in the menu bar. 2. Open the Days folder in your chosen save location. 3. Search the markdown files with grep for a phrase you remember seeing. 4. Open the file path or URL shown in the matching block. ### Prompt to paste into your agent ```text In the dragthelake/ambient-context record folder, search the Days files for and list each match with its app, time and any file path or URL. Expected output: a list of matches with timestamps. Verify that each listed path or URL opens the original document. ``` ### Caveat References appear only where the app exposes a document path or URL, so some blocks will show text alone. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/dragthelake-ambient-context/find-page-or-file-seen-earlier/ ## How to build a cited knowledge base of people and commitments with Ambient Context Repo: [dragthelake/ambient-context](https://github.com/dragthelake/ambient-context) | Difficulty: Advanced | Time: 1 hour | Basis: From the README Get six cited pages on people, commitments, threads, products, issues and reading from one day. **Who it is for:** A project lead who needs a running memory of who said what and which commitments were made across a day of messages and tickets. **The problem:** Commitments and names made in passing get lost between calls, chats and emails, and nobody writes them down. **The outcome:** Knowledge pages in KB// where each claim points back to a block in the record. ### Steps 1. Record the day, then open the Agent tab and connect an agent CLI you already have. 2. Press Process day on the recorded date, or let the scheduled time run it. 3. Open KB//people.md and commitments.md and check the citations. 4. Open KB//manifest.md to confirm which pages were written. ### Prompt to paste into your agent ```text Using the dragthelake/ambient-context knowledge folder, read KB//commitments.md and list each commitment with its owner and cited block. Expected output: a table with columns commitment, owner and source block. Verify that every row has a citation that exists in Days//. ``` ### Caveat Knowledge is built by the agent CLI you connect, which runs under your own subscription, so its quality depends on that tool. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/dragthelake-ambient-context/build-cited-knowledge-base-of-people/ ## How to tighten capture rules after reviewing a day with Ambient Context Repo: [dragthelake/ambient-context](https://github.com/dragthelake/ambient-context) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Use the record to set rules that stop capturing apps, sites or text you do not want kept. **Who it is for:** A privacy-minded user who spots something sensitive in the day's record and wants to stop it being recorded again. **The problem:** You may only notice a sensitive page after it has already been written to the folder. **The outcome:** Capture rules that exclude that app or site, or redact that kind of text, from the next recording. ### Steps 1. Open the day in the Context tab and find the block you want to restrict. 2. Choose Never record this app, Headings only for this site, or Redact text like this on that block. 3. Optionally install the bundled skills from Settings and ask the agent to use the tune rules skill. 4. Check the next recording to confirm the rule took effect. ### Prompt to paste into your agent ```text In the dragthelake/ambient-context app, use the tune rules skill to review the Context records for and propose capture rules for apps or sites that should not be kept. Expected output: a short list of proposed rules, each naming the app or site and the action. Verify each proposal against the block it came from before applying it. ``` ### Caveat These are heuristics rather than a guarantee, so keep treating the folder as sensitive even after adding rules. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/dragthelake-ambient-context/tighten-capture-rules-after-review/