restate
restate to say the last message again in plain words, catch the user up since their last message, and summarize any decision they need to make.
Skill by prathamdby
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/prathamdby-restate
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 25, 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/restate
What's inside
- Skill name
restate- Description
- restate to say the last message again in plain words, catch the user up since their last message, and summarize any decision they need to make.
- License
- MIT
- Instructions
- 51 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Restate ## Options The source is the agent's previous user-visible message in this conversation. The window is every turn after the user's previous message through that reply. "Restate", "say that again simply", "catch me up", and "what do I need to decide" select this procedure. A request to keep the jargon, skip the catch-up, rewrite some other text, or change the underlying work blocks. No previous message blocks. Record `source | window | decisions | terminal`. ## Voice Write as one person talking to another. Use everyday words and short sentences. Keep the facts, names, and outcomes. Say each thing once, in fewer words than a long source. Use a specialist term only when the user used it or the fact disappears without it. ## 1. Gather Read the source message and the window. Fix what changed, what finished, what is still open, and every action or decision waiting on the user. Use only facts the conversation supports. Done when the source, window, and user decisions are fixed, or the run is blocked. ## 2. Write The reply has this order: 1. Restate the source so a reader who missed it can follow it alone. 2. Catch up on the window: what changed, what finished, and what is still open. When that repeats the restatement, one placing sentence is enough. 3. When the user has an action or decision, close with an executive summary: the choice, the facts that matter, and what each option leads to. Several decisions each get that context. No user decision means no summary. Done when the reply is plain, complete for the window, and every user decision can be made from the summary. ## 3. Deliver Send the reply in the current conversation. Leave files, tasks, and git unchanged. Done when the reply is sent. Terminals: RESTATED, BLOCKED.
From prathamdby/skills, MIT.
More from prathamdby/skills
- 0.621score
autoplan
prathamdby
autoplan to produce a grounded implementation plan without questions; gated design uses upfront-design.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
box
prathamdby
box to manage external Git clones and answer questions from their local source.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
claude-code
prathamdby
claude-code to run the local Claude CLI or manage its installation, auth, MCP, and plugins.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
codex
prathamdby
codex to run the local Codex CLI or manage its installation, login, MCP, and plugins.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
cursor-agent
prathamdby
cursor-agent to run the local Cursor CLI or manage installation, login, integrations, and workers.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
fix-pr
prathamdby
fix-pr to handle PR feedback and blocking CI in one pass; invokes gh and commit.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
gh
prathamdby
gh for PR state, discussions, CI diagnosis, replies, and nontrivial GitHub commands; fix-pr invokes it.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
handoff
prathamdby
handoff to save resumable context or continue from a validated handoff document.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
make-pr
prathamdby
make-pr to publish committed changes and create or update their pull request.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
recon
prathamdby
recon to read a saved codebase map, map the current repo, or refresh it after changes.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
upfront-design
prathamdby
upfront-design to agree phased designs before coding, resume drafts, or verify approved milestones.
SkillAFreshness: A★ 6Claude Code +9 - 0.616score
todoist-task
prathamdby
todoist-task to create or preview self-contained tasks with requested metadata.
SkillAFreshness: A★ 6Claude Code +9
Signals
- Stars
- 6
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 11, 2026 (18 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/prathamdby/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from prathamdby
- 0.621score
box
prathamdby
box to manage external Git clones and answer questions from their local source.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
claude-code
prathamdby
claude-code to run the local Claude CLI or manage its installation, auth, MCP, and plugins.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
codex
prathamdby
codex to run the local Codex CLI or manage its installation, login, MCP, and plugins.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
cursor-agent
prathamdby
cursor-agent to run the local Cursor CLI or manage installation, login, integrations, and workers.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
fix-pr
prathamdby
fix-pr to handle PR feedback and blocking CI in one pass; invokes gh and commit.
SkillAFreshness: A★ 6Claude Code +9 - 0.621score
autoplan
prathamdby
autoplan to produce a grounded implementation plan without questions; gated design uses upfront-design.
SkillAFreshness: A★ 6Claude Code +9
Cite this page
restate skill. Agent Mart. https://agentmart.dev/skills/prathamdby-restate (last updated 2026-10-24)