recap
Skill0 starsChanged 2 months agoMITView on GitHub
Summarize all user requests from the current AI chat session and produce a concise rich HTML recap (requested vs delivered). Use when the user invokes /recap or asks for a session summary, delivery recap, or handoff document.
- Reads credentials
45 other projects use a version of this.
Adopted by
Most recently taken up by
- web3illy/private-second-brain-skillstheir own version, first seen Nov 16, 2026
- Integralist/agent-skillstheir own version, first seen Nov 14, 2026
- mel0nyrame/skillstheir own version, first seen Nov 8, 2026
- delorenj/skillstheir own version, first seen Nov 7, 2026
- acegalaxy-co/ace_commons-ccswitch-clitheir own version, first seen Nov 1, 2026
- kanalias/ace_commons-ccswitch-clitheir own version, first seen Nov 1, 2026
Used with
Projects with it also use glossary, orchestrate and 7 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
Microsoft AzureSkillglossary
Short glossary of domain terms used by the team.
Also in 3 projects that use this
varun29ankuSSkillorchestrate
This skill should be used when the user asks to 'orchestrate a task', 'break down work into parallel agents', 'coordinate subtasks', 'run agents in parallel', or mentions 'multi-agent'. Decomposes complex tasks into tracked subtasks, dispatches parallel subagents, and coordinates until completion.
Also in 4 projects that use this
sunnyclinches44Skillexplain
Explains code using real-world analogies and ASCII diagrams
Also in 3 projects that use this
Microsoft AzureSkillbuild
Build the TypeSpec Rust emitter from source. Run this after cloning the repo, after pulling changes, or when TypeScript source files have been modified.
Also in 3 projects that use this
nicolasestremSkillbrainstorm
Structured idea generation and solution exploration with systematic analysis.
Also in 3 projects that use this
BlockSkillplan
Draft a TDD, PRD, or lightweight plan in the conversation without committing it.
Also in 3 projects that use this
Other versions45 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 45
- first-fluke/fullstack-starter★ 226same name
- backnotprop/bro★ 182same name
- benoror/obsidianos_work★ 166same name
- Soul-Brews-Studio/arra-oracle-skills-cli★ 121same name
- mbwsims/claude-universe★ 71same name
- Integralist/agent-skills★ 29same name
- TakaGoto/rag-learning-academy★ 24same name
- delorenj/skills★ 17same name
- ivklgn/ai-kit★ 12same name
- ivklgn/bag-of-skills★ 12same name
35 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository26 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .cursor/rules/automate-before-manual.mdcCursor rule
- .cursor/rules/finish-task-dev-server.mdcCursor rule
- brainstormingSkill
- caveman-commitSkill
- find-skillsSkill
- semantic-versionSkill
- systematic-debuggingSkill
- verification-before-completionSkill
- writing-plansSkill
- ask-before-architectureSkill
- automate-before-manualSkill
- caveman-commitSkill
- diagrams-mermaidSkill
- dont-be-lazySkill
- dont-forgetSkill
- find-jobSkill
- finish-with-dev-serverSkill
- google-calendar-mcpSkill
- init-fmzSkill
- instagram-chrome-postSkill
- project-initSkill
- semantic-versionSkill
- skills-sh-maintainerSkill
- ui-change-e2eSkill
- validate-before-shareSkill
The file
We split this skill into the rules it gives an agent
12 rules. 0 are common, 7 are distinctive, and 5 are very specific to this project.
0
Common
Most of the 39 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 28
Ensure local dev server: when the project rules require it (e.g.
- Very specific: line 29
Read repo docs for start commands (TerapIA: npm run db:up + npm run dev → default http://localhost:3000).
- Distinctive: line 30
If down, start it before finishing the recap.
- Distinctive: line 38
Before generating HTML, apply anti-ui-slop and frontend-design:
- Very specific: line 40
No generic slop: avoid cream #F4F1EA templates, default Inter/system-only stacks, and numbered cards without meaning.
- Very specific: line 42
Ground in the product: when the repo has design tokens (e.g.
- Very specific: line 58
Use a stable kebab-case data-topic-id (slug from title).
- Distinctive: line 79
Keep copy short; max ~8–12 topic cards.
- Distinctive: line 81
Use this dark token set when no project tokens exist:
- Distinctive: line 125
Accurate: only claim delivery for work that was implemented or explicitly output.
- Very specific: line 126
No secrets: never include tokens, passwords, or .env values.
- Distinctive: line 130
If the recap reveals unfinished work, add a Pending section at the bottom (plain list).
“Common” is measured: kept by at least 6 in 10 of the 39 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/recap — Session delivery recap
Produce a single self-contained HTML file that recaps the conversation: what the user asked for (brief paraphrase) and what was actually delivered.
When to use
- User invokes
/recap - User asks for session summary, recap, handoff, or "what was asked vs delivered"
- End of a long multi-topic agent session before closing
Workflow
- Read the full conversation (all user messages, chronological). Group related turns into topics; do not list every micro-message.
- For each topic, write at most two short blocks:
- Requested — user intent in plain language (1–2 sentences). Use the project's UI language (Portuguese for TerapIA unless the project is English-only).
- Delivered — concrete outcomes: files, routes, scripts, rules, tests; include paths when helpful.
- Skip failed or abandoned attempts unless the user should know about them.
- Generate HTML using the template below (inline CSS only, no external dependencies).
- Ensure local dev server when the project rules require it (e.g.
AGENTS.md, user rules):- Read repo docs for start commands (TerapIA:
npm run db:up+npm run dev→ defaulthttp://localhost:3000). - If down, start it before finishing the recap.
- Validate HTTP 200 on health or main route; record the effective URL (port may differ).
- Include test login hints only if documented in the repo (never secrets from
.env).
- Read repo docs for start commands (TerapIA:
- Save to
docs/session-recaps/YYYY-MM-DD-<slug>.htmlwhen the project has adocs/folder; otherwisesession-recap.htmlat workspace root or a path the user specifies. - Reply with the file path, local URL, and a short markdown table summary (optional).
UI design (required)
Before generating HTML, apply anti-ui-slop and frontend-design:
- No generic slop: avoid cream
#F4F1EAtemplates, default Inter/system-only stacks, and numbered cards without meaning. - Dark mode by default for recap HTML (
color-scheme: dark,meta color-scheme). - Ground in the product: when the repo has design tokens (e.g. TerapIA
dashboard-theme-presets.ts→petroleo-atual), reuse primary, bg, ink, muted — do not invent a unrelated palette. - Typography: pair a restrained serif display (e.g. Literata) with a clean sans body (e.g. IBM Plex Sans); not Arial/Inter alone.
- Hierarchy: eyebrow → title → local test panel (pill link grid + health dot) → clickable topic cards (mark as seen) → pending footer.
- Accessibility: visible
:focus-visible,prefers-reduced-motion, sufficient contrast on dark surfaces.
Topic cards — marcar como visto (required)
Each topic card is a conversation chunk the user can click to mark as read. State persists in localStorage (per file path + topic id).
- Markup — every topic:
Use a stable kebab-case<article class="topic" data-topic-id="slug-curto-do-topico"> <h2>Título do tópico</h2> ... </article>data-topic-id(slug from title). Required for persistence. - Progress chip in header stats:
<span class="chip chip--progress" data-recap-progress hidden>0/0 vistos</span> - CSS — include styles from
reference/topic-seen.css(or equivalent inline). Seen state:.topic--seen— lower opacity, desaturated, muted left bar, "Visto" pill top-right. - JS — before
</body>, inline the script fromreference/topic-seen.js:- Click or Enter/Space toggles seen
role="button",aria-pressed,tabindex="0"- Storage key:
recap-seen:{pathname}:{data-topic-id} - Updates
[data-recap-progress]counter
- UX copy (optional hint below topics title): Clique em um card para marcar como visto.
Reference files (same folder as this skill): reference/topic-seen.css, reference/topic-seen.js.
HTML template
Keep copy short; max ~8–12 topic cards. Label text follows project locale (PT-BR: Pedido / Entregue / Pendências).
Use this dark token set when no project tokens exist:
| Token | Default |
--bg-page | #102223 |
--bg-card | #16302f |
--primary | #2d6a6b |
--secondary | #7fb5b0 |
--ink | #f2f5f4 |
--muted | #9bb0ae |
Structure (inline CSS + optional Google Fonts):
<!DOCTYPE html>
<html lang="pt-BR">
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1" />
<meta name="color-scheme" content="dark" />
<title>Recap — {Project} — {Date}</title>
<!-- Literata + IBM Plex Sans or project-appropriate pair -->
<style>/* dark tokens, radial glow on body::before, card shadows */</style>
</head>
<body>
<div class="wrap">
<p class="eyebrow">{Project} · sessão agente</p>
<header>...</header>
<section class="local-env"><!-- pill link grid + health dot when dev server required --></section>
<div class="topics">
<p class="topics-hint">Clique em um card para marcar como visto.</p>
<article class="topic" data-topic-id="example-topic">...</article>
</div>
<section class="pending">...</section>
<footer>Generated by /recap</footer>
</div>
<script>/* inline topic-seen.js */</script>
</body>
</html>Quality bar
- Concise: each topic fits one card; no code dumps in HTML.
- Accurate: only claim delivery for work that was implemented or explicitly output.
- No secrets: never include tokens, passwords, or
.envvalues.
Optional follow-up
If the recap reveals unfinished work, add a Pending section at the bottom (plain list).
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

