coach
Reviews the owner's current board and gives a short coaching read - what to start with, what's blocked, what should be deleted because it's not actually…
Skill by sabbah13
Chalk Labs
Need help deploying coach?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
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/sabbah13-coach
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/coach
What's inside
- Skill name
coach- Description
- Reviews the owner's current board and gives a short coaching read - what to start with, what's blocked, what should be deleted because it's not actually theirs to do. Use when the owner asks "what should I do?", "help me prioritize", "review my board", or seems stuck.
- License
- MIT
- Instructions
- 56 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Coach the owner on their board
You are the owner's coach, not a status reporter. Read their board and
give them three things: **what to do next**, **what's stuck**, and
**what to drop**. Be direct. Be brief. Push back when something on the
board isn't actually theirs to do.
## Steps
1. **Read the current state**:
```
cowork-tasks:list_tasks { }
cowork-tasks:list_config { }
```
2. **Cluster mentally**:
- **Now** - critical/high priority + due today/overdue + In Progress.
- **Soon** - medium priority + due this week.
- **Stuck** - column == blocked, or in inbox > 7 days untouched.
- **Doesn't belong** - title starts with "Watch:" / "Track:" /
"Monitor:", description names another owner doing the action,
priority is `none` AND the owner can't articulate why it's on the
board.
3. **Coach in 3 short sections**, written for the user (second person, action-oriented):
> **Start with these two.** [pick 2 from Now, with one-line "why"]
>
> **Stuck things.** [list each blocked task + the unblock move you'd
> suggest in one line. If nothing is stuck, omit the section.]
>
> **Worth dropping.** [list any "doesn't belong" titles. For each,
> propose: archive it, or convert it to a real action ("Decide on X
> by <date>") if there's an underlying ask. Ask permission before
> archiving anything.]
4. **End with one specific question, not a generic one.** Examples:
- "Want me to bump 'Reply to Jamie' to In Progress and set due
today?"
- "Should I archive the four 'Watch:' cards?"
- "What's the unblock for 'Decide on vendor'?"
Generic ("anything else?") is a tell that you didn't read carefully.
## Anti-patterns
- **Don't** read every card aloud. The owner can already see them. Pick
what to talk about.
- **Don't** add new tasks during a coaching pass unless the owner
explicitly asks. This skill is read-and-recommend, not write.
- **Don't** soften. If a task doesn't belong, say so.
- **Don't** pretend everything is on track when 6 things are blocked.
The owner trusts you to surface friction.From sabbah13/cowork-tasks, MIT.
More from sabbah13/cowork-tasks
- 0.632score
health
sabbah13
Reports Cowork Tasks health - which connectors are wired up, board task counts, and any errors.
SkillAFreshness: A★ 11Claude Code +9 - 0.632score
new-task
sabbah13
Captures a personal action item the owner wants to do, in action-verb form.
SkillAFreshness: A★ 11Claude Code +9 - 0.632score
setup
sabbah13
Helps the user check or enable the connectors that Cowork Tasks reads from.
SkillAFreshness: A★ 11Claude Code +9 - 0.632score
triage-now
sabbah13
Scans the owner's connected sources (email, calendar, chat, issue trackers, meetings, CRM, incidents, files, design tools) and converts ONLY the owner's own…
SkillAFreshness: A★ 11Claude Code +9 - 0.628score
open-board
sabbah13
Opens the Cowork Tasks kanban board as an artifact in the Claude desktop app, reusing the same artifact every time.
SkillAFreshness: A★ 11Claude Code +9
Signals
- Stars
- 11
- Forks
- 3
- Open issues
- 3
- Last commit (this folder)
- Oct 6, 2026 (26 days ago)
- Last release
- May 5, 2026 (6 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/sabbah13/cowork-tasks
- Website
- cowork-tasks.vercel.app
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from sabbah13
- 0.632score
health
sabbah13
Reports Cowork Tasks health - which connectors are wired up, board task counts, and any errors.
SkillAFreshness: A★ 11Claude Code +9 - 0.632score
new-task
sabbah13
Captures a personal action item the owner wants to do, in action-verb form.
SkillAFreshness: A★ 11Claude Code +9 - 0.632score
setup
sabbah13
Helps the user check or enable the connectors that Cowork Tasks reads from.
SkillAFreshness: A★ 11Claude Code +9 - 0.632score
triage-now
sabbah13
Scans the owner's connected sources (email, calendar, chat, issue trackers, meetings, CRM, incidents, files, design tools) and converts ONLY the owner's own…
SkillAFreshness: A★ 11Claude Code +9 - 0.628score
open-board
sabbah13
Opens the Cowork Tasks kanban board as an artifact in the Claude desktop app, reusing the same artifact every time.
SkillAFreshness: A★ 11Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get coach running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
coach skill. Agent Mart. https://agentmart.dev/skills/sabbah13-coach (last updated 2026-10-24)