summary
Give the user a fast, ADHD/dyslexia-friendly recap of the current work or conversation — the outcome first, then what (if anything) you need from them.
Skill by kirinnee
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/kirinnee-summary
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 26, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/summary
What's inside
- Skill name
summary- Description
- Give the user a fast, ADHD/dyslexia-friendly recap of the current work or conversation — the outcome first, then what (if anything) you need from them. Use when running /summary, or when asked to "summarize", "recap", "where are we", "catch me up", or "what do you need from me".
- License
- MIT
- Instructions
- 71 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Summary — Fast, Readable Recap Produce a recap the user can read in **under 15 seconds**. The reader has ADHD and dyslexia: lead with the answer, keep it short, make any action they must take impossible to miss. This is a **communication skill** — it changes how you write the recap, not the underlying work. Do not start new tasks, edit code, or run long commands. Gather just enough state to be accurate, then write. ## When to Use - Running `/summary` - "Summarize / recap / TL;DR this" - "Where are we?" / "Catch me up" / "What's the status?" - "What do you need from me?" - After finishing a chunk of work, before handing back to the user ## What to Summarize Recap the **current session/work**: what was done, where things stand now, and what's blocked or pending. If the user gives a focus argument, narrow to that. If you genuinely lack context (fresh session, nothing done yet), say so in one line and ask what they want summarized — don't invent a recap. ## Output Format Keep it tight. Target **under ~120 words**. Use this shape: ``` **<one-line outcome — the single most important thing>** What's done: - <short point> - <short point> Need from you: - <action 1 — concrete and obvious> (or: "Nothing — all good ✅") ``` Rules for the writing: - **Lead with the outcome.** First line = the answer/state, bolded. Details after. - **One idea per bullet.** Short lines. No walls of text, no nested sub-bullets. - **Plain words.** Avoid jargon; if a term is unavoidable, give a 3-word gloss. - **Make the ask loud.** The "Need from you" block always comes last and is always present — even if it's just "Nothing needed." - **Bold sparingly.** Only the outcome line (and maybe one key word). If everything is bold, nothing stands out. - **Cut filler.** No preamble like "Here's a summary of…". Just give it. - Drop the "What's done" section if there's nothing meaningful to list — but never drop "Need from you". ## The "Need from You" Block — Always Include It This is the point of the skill: the user must instantly see what's on them. For each item, say the **concrete action**, not vague status. Good: "Approve the migration plan", "Pick: rebase or merge?", "Run `hms` to apply". Bad: "There are some pending decisions." If nothing is needed, say exactly that — a clear "Nothing from you — done ✅" so they can relax. ## Example ``` **Auth refactor is done and tests pass — ready for your review.** What's done: - Moved login to the new token flow - Updated 4 call sites + added tests (all green) Need from you: - Review the PR, then merge it yourself when happy - One open question: keep the old `/login` route as a redirect? (y/n) ``` ## Notes - Accuracy first: a fast recap that's wrong is worse than a slightly longer one that's right. Check real state (`git status`, what you actually changed) before claiming "done". - This format applies to the recap only — it does not change commit messages, code, or file contents.
From kirinnee/home-manager, MIT.
More from kirinnee/home-manager
- 0.609score
create-skill
kirinnee
Create a new agent skill in the open SKILL.md standard (works with Claude Code AND Codex).
SkillAFreshness: A★ 4Claude Code +9 - 0.609score
floop
kirinnee
Iteratively review-and-resolve the current changes using parallel subagent reviewers across multiple lenses (bugs, quality, completion), looping until every…
SkillAFreshness: A★ 4Claude Code +9 - 0.609score
babysit-pr
kirinnee
Open a pull request for the current branch and babysit it to ready-to-merge: push, create/reuse the PR, then loop — fix failing CI, keep the branch current…
SkillAFreshness: A★ 4Claude Code +9 - 0.609score
clickup-task
kirinnee
Create tasks in ClickUp with support for status changes, dependencies, assignees, and tags.
SkillAFreshness: A★ 4Claude Code +9 - 0.609score
liftoff-kubernetes
kirinnee
Kubernetes cluster access via loctl (kubectl, helm, stern).
SkillAFreshness: A★ 4Claude Code +9 - 0.609score
liftoff-observability
kirinnee
Observability via loctl (promtool, grafanactl, logs).
SkillAFreshness: A★ 4Claude Code +9 - 0.609score
liftoff-ops
kirinnee
Umbrella entry point for Liftoff infrastructure access via loctl: Kubernetes, Mimir, Loki, Grafana, ArgoCD, AWS, Helm, Terraform.
SkillAFreshness: A★ 4Claude Code +9 - 0.608score
liftoff-argocd
kirinnee
ArgoCD via loctl.
SkillAFreshness: A★ 4Claude Code +9 - 0.608score
liftoff-aws
kirinnee
AWS and Terraform via loctl.
SkillAFreshness: A★ 4Claude Code +9 - 0.584score
frontend-design
kirinnee
Create distinctive, production-grade frontend interfaces with high design quality.
SkillAFreshness: A★ 4Claude Code +9
Signals
- Stars
- 4
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 20, 2026 (8 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/kirinnee/home-manager
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from kirinnee
- 0.609score
create-skill
kirinnee
Create a new agent skill in the open SKILL.md standard (works with Claude Code AND Codex).
SkillAFreshness: A★ 4Claude Code +9 - 0.609score
floop
kirinnee
Iteratively review-and-resolve the current changes using parallel subagent reviewers across multiple lenses (bugs, quality, completion), looping until every…
SkillAFreshness: A★ 4Claude Code +9 - 0.609score
babysit-pr
kirinnee
Open a pull request for the current branch and babysit it to ready-to-merge: push, create/reuse the PR, then loop — fix failing CI, keep the branch current…
SkillAFreshness: A★ 4Claude Code +9 - 0.609score
clickup-task
kirinnee
Create tasks in ClickUp with support for status changes, dependencies, assignees, and tags.
SkillAFreshness: A★ 4Claude Code +9 - 0.609score
liftoff-kubernetes
kirinnee
Kubernetes cluster access via loctl (kubectl, helm, stern).
SkillAFreshness: A★ 4Claude Code +9 - 0.609score
liftoff-observability
kirinnee
Observability via loctl (promtool, grafanactl, logs).
SkillAFreshness: A★ 4Claude Code +9
Cite this page
summary skill. Agent Mart. https://agentmart.dev/skills/kirinnee-summary (last updated 2026-10-24)