Agent skill

Ambient Context

by dragthelake in dragthelake/ambient-context

A skill your agent uses when the user refers to what they were working on, asks to catch up, mentions their day, week, record or Ambient Context, or when a task needs to know what happened recently…

Apache-2.0Auto-check passedAgent Workflows

Install Ambient Context

skills CLI
$ npx skills add dragthelake/ambient-context --skill ambient-context -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install dragthelake/ambient-context ambient-context --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/dragthelake/ambient-context.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/ambient-context .claude/skills/ambient-context && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
ambient-context
GitHub stars
206
Token cost
~985 tokens
SKILL.md length
570 words
Files
1
Skills in repo
5
Repo updated
First seen
Licence
Apache-2.0

At a glance

A skill your agent uses when the user refers to what they were working on, asks to catch up, mentions their day, week, record or Ambient Context, or when a task needs to know what happened recently…

  • The user refers to what they were working on
  • SKILL.md covers The record, Which tool, Restraint and Privacy, plus 1 more section
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
  • Asks to catch up

What it does

Ambient Context is an agent skill from dragthelake/ambient-context. Use when the user refers to what they were working on, asks to catch up, mentions their day, week, record or Ambient Context, or when a task needs to know what happened recently on this Mac. Reads the Ambient Context record through its MCP tools and says how much of it to read.

Its SKILL.md is about 990 tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Agent Workflows, covering MCP servers. The repository describes itself as: A menu bar app that keeps a written record of what you worked on. The licence is Apache-2.0.

When your agent uses it

  • The user refers to what they were working on
  • Asks to catch up
  • Mentions their day
  • Ambient Context

Example prompts

  • “/ambient-context”

What it can do on your machine

Read from SKILL.md and the folder at commit 7e57155. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Ambient Context loads about 985 tokens when it runs. Until then it costs about 74 tokens; SKILL.md has 570 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~74
When it runs · the whole SKILL.md, loaded when a task matches
~985

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from dragthelake/ambient-context at commit 7e57155, republished under its Apache-2.0 licence (© dragthelake). 570 words, ~985 tokens.

Download SKILL.mdSave it as .claude/skills/ambient-context/SKILL.md (or your agent's skills folder).
name
ambient-context
description
Use when the user refers to what they were working on, asks to catch up, mentions their day, week, record or Ambient Context, or when a task needs to know what happened recently on this Mac. Reads the Ambient Context record through its MCP tools and says how much of it to read.
license
Apache-2.0
metadata.tools
capture_status list_days read_summary read_day read_kb search_record read_ledger summarise_day ingest_day add_rule update_rule remove_rule set_prompt set_config

Ambient Context

Ambient Context is a macOS menu bar app that records what its user works on. Every few seconds it reads the text of the focused window and appends it to a plain-text record, one folder per day, on the user's own disk. An agent the user runs later turns each day into a summary and a small knowledge base. The app is also an MCP server named ambient-context, and that server is how you reach the record.

The record

For each day there is:

  • Three raw files. apps (everything not routed elsewhere), websites (a table of pages visited with dwell time) and messages (chat and mail). Blocks are headed ## HH:MM–HH:MM · App · Window title. A block body holds only lines new to that day, so a bare heading means the user was there, looking at things already recorded. The text is an accessibility-tree scrape, not prose: good for what was on screen, silent about the rest. [redacted] marks a scrubbed secret.
  • A summary. One distilled account of the day, written by the user's agent, with time-range citations.
  • A knowledge base. Six files (people, commitments, threads, products, issues, reading), every line cited.
  • A ledger. Every model action and every configuration change, with who made it. Your writes appear here under this client's name.

Which tool

You wantCall
What existslist_days
What a day was aboutread_summary first; read_kb for people, commitments and threads
Detail on a window of timeread_day with from and to, and file when it is not apps
When something happenedsearch_record
Whether recording is on and what is runningcapture_status, and only when that is the question
What changed and who changed itread_ledger

A day with no summary can be built with summarise_day and ingest_day. Both take minutes and run the user's own agent on their machine, so say that you are starting one, and do not start a second while one runs.

Show full SKILL.md (251 more words)Show less

Restraint

  • Read the narrowest thing that answers the question: a summary before a day file, a time window before a whole day, a search before a read.
  • Quote a few lines when they carry the answer. Never paste a day file, a summary or a block wholesale into a reply.
  • A record that does not contain the thing is a normal result. Say so plainly and do not fill the gap with a guess.
  • Reading needs nothing running. Writing (capture, rules, the prompt, config) needs the app open, and every write is logged with your name on it. After a write, say what you changed in one line.
  • Do not call add_rule, update_rule, remove_rule, set_prompt or set_config unless the user asked for that change. The ambient-context-tune-rules skill handles rule changes and asks first.

Privacy

The record can hold messages, page contents and documents the user never meant to share. It stays on this machine. Do not send it to another service, write it into files in a repository, paste it into an issue or a commit, or summarise it somewhere other people read, without asking first and naming what would leave.

When the server is not there

If the tools are missing, the server is not registered: ask the user to open Ambient Context, Settings, MCP server. If a write returns not_running, the app is closed: reads still work and writes will once it is open. Say which of the two it is and stop. Do not guess at the record.

© dragthelake, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/ambient-context of dragthelake/ambient-context.

Open the folder on GitHubat commit 7e57155

Compare with similar skills

Ambient Context next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Ambient Context compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Ambient Context this skilldragthelake/ambient-context206—~985Automated safety check: PassApache-2.0
MCP Server Builderanthropics/skills180k63 repos~2.3kAutomated safety check: PassApache-2.0
MCP Server BuildershareAI-lab/learn-claude-code78k4 repos~1.2kAutomated safety check: PassMIT
MCP Integration for Pluginsanthropics/claude-plugins-official38k11 repos~3.1kAutomated safety check: PassApache-2.0
Microsoft Skill CreatorMicrosoftDocs/mcp1.9k3 repos~2.1kAutomated safety check: PassCC-BY-4.0
Crush Configurationcharmbracelet/crush29k—~3.7kAutomated safety check: PassCustom licence

Similar skills

  • MCP Server Builder

    anthropics/skills

    Official

    Guides the design and implementation of Model Context Protocol servers in TypeScript or Python, from tool naming and error messages to evaluation.

    180k GitHub starsUsed in 63 repos~2.3k tokens
    Agent WorkflowsAuto-check passed
  • MCP Server Builder

    shareAI-lab/learn-claude-code

    Walks through building MCP servers in Python or TypeScript that expose tools, resources and prompts to Claude, with templates, registration and testing.

    78k GitHub starsUsed in 4 repos~1.2k tokens
    Agent WorkflowsAuto-check passed
  • MCP Integration for Plugins

    anthropics/claude-plugins-official

    Official

    Explains how to bundle Model Context Protocol servers in a Claude Code plugin, covering config files, stdio, SSE, HTTP and WebSocket server types, and authentication.

    38k GitHub starsUsed in 11 repos~3.1k tokens
    Agent WorkflowsAuto-check passed
  • Microsoft Skill Creator

    MicrosoftDocs/mcp

    Official

    Create agent skills for Microsoft technologies using official documentation.

    1.9k GitHub starsUsed in 3 repos~2.1k tokens
    Agent WorkflowsAuto-check passed
  • Crush Configuration

    charmbracelet/crush

    Explains how to configure the Crush coding agent with crushrc or crush.json, covering providers, models, LSPs, MCP servers, hooks, permissions and config precedence.

    29k GitHub stars~3.7k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Context Mode Output Sandbox

    mksglu/context-mode

    Routes large command, file, API and browser output through context-mode tools so only the needed result enters the agent's context, instead of dumping it via Bash.

    26k GitHub stars~4.1k tokensUpdated today
    Agent WorkflowsAuto-check passed

More from dragthelake/ambient-context

  • Ambient Context Tune Rules

    dragthelake/ambient-context

    A skill your agent uses when the user says something should not be captured, is landing in the wrong file, keeps showing up in summaries, or asks to tidy or review their Ambient Context rules.

    206 GitHub stars~496 tokensUpdated yesterday
    Auto-check passed
  • Ambient Context Catch Me Up

    dragthelake/ambient-context

    A skill your agent uses when the user asks what they were doing, where they got to, what is still open, or to be caught up after a break.

    206 GitHub stars~408 tokensUpdated yesterday
    Auto-check passed
  • Ambient Context Standup

    dragthelake/ambient-context

    A skill your agent uses when the user asks for a standup, daily update, status for a channel, or what to say they did.

    206 GitHub stars~343 tokensUpdated yesterday
    Auto-check passed
  • Ambient Context Weekly Review

    dragthelake/ambient-context

    A skill your agent uses when the user asks for a weekly review, a week note, what happened this week, or what they spent the week on.

    206 GitHub stars~362 tokensUpdated yesterday
    Auto-check passed

Categories

Questions about Ambient Context

What does Ambient Context do?

A skill your agent uses when the user refers to what they were working on, asks to catch up, mentions their day, week, record or Ambient Context, or when a task needs to know what happened recently…. Ambient Context is an agent skill from dragthelake/ambient-context. Use when the user refers to what they were working on, asks to catch up, mentions their day, week, record or Ambient Context, or when a task needs to know what happened recently on this Mac.

When should I use Ambient Context?

Ambient Context fits situations like: the user refers to what they were working on; asks to catch up; mentions their day; ambient Context.

How do I install Ambient Context in Claude Code?

Run `npx skills add dragthelake/ambient-context --skill ambient-context -a claude-code`. Or copy the skill folder (skills/ambient-context in dragthelake/ambient-context) into .claude/skills/ambient-context in your project. Claude Code loads it when a task matches its description.

How do I install Ambient Context in Codex?

Run `npx skills add dragthelake/ambient-context --skill ambient-context -a codex`. Or copy the skill folder (skills/ambient-context in dragthelake/ambient-context) into .agents/skills/ambient-context in your project. Codex loads it when a task matches its description.

Can I use Ambient Context in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add dragthelake/ambient-context --skill ambient-context -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ambient-context, .gemini/skills/ambient-context, .github/skills/ambient-context and .opencode/skills/ambient-context in your project.

What does Ambient Context need to run?

SKILL.md names no scripts, command-line tools or credentials: Ambient Context is instructions for the agent only.

Does Ambient Context access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Ambient Context safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Ambient Context use?

Ambient Context is published under the Apache-2.0 licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Ambient Context use?

About 985 tokens (SKILL.md is roughly 3.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Ambient Context?

Skills that share tags, products or a category with Ambient Context: MCP Server Builder (anthropics/skills, 180k stars), MCP Server Builder (shareAI-lab/learn-claude-code, 78k stars), MCP Integration for Plugins (anthropics/claude-plugins-official, 38k stars) and Microsoft Skill Creator (MicrosoftDocs/mcp, 1.9k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Ambient Context?

dragthelake (a GitHub user) maintains it in dragthelake/ambient-context, which has 206 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on October 8, 2026.

Source: dragthelake/ambient-context on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.