Skillquality 0.47

running-effective-meetings

Run effective meetings: brief, agenda, facilitation script, decision log. See also: running-effective-1-1s (1:1s), running-offsites (offsites).

Price
free
Protocol
skill
Verified
no

What it does

Running Effective Meetings

Scope

Covers

  • Deciding whether a meeting is needed (vs async doc/message)
  • Choosing the right meeting format (operational / strategic / decision / workshop)
  • Preparing a pre-read and “priming” to avoid context-less discussions
  • Facilitating discussion, capturing decisions, and ensuring follow-through
  • Improving recurring meeting hygiene (split/kill/shorten meetings)

When to use

  • “Create an agenda and facilitation plan for a decision meeting.”
  • “We need a weekly operational cadence — design the meeting format and templates.”
  • “This meeting keeps going in circles — redesign it so we decide and assign actions.”
  • “Turn this messy invite into a clear meeting brief + pre-read + follow-up.”

When NOT to use

  • The goal is purely status updates (use async updates instead).
  • There is no discussion to be had (you already have the decision and only need to broadcast it).
  • The “decision owner” is not attending and cannot delegate decision rights.
  • You need a recurring 1:1 with a direct report or manager (use running-effective-1-1s; different structure and relationship dynamics).
  • You need a multi-day offsite or retreat (use running-offsites; this skill covers single meetings, not multi-session programs).
  • You need a formal decision process with RACI, options scoring, and escalation (use running-decision-processes; this skill captures decisions within meetings but does not design the full decision framework).
  • You need a design review or architecture review with technical critique (use running-design-reviews; specialized facilitation and feedback norms apply).
  • The topic is a sensitive HR/legal/medical issue requiring specialist handling.

Inputs

Minimum required

  • Meeting objective (what changes after the meeting)
  • Meeting type (or “recommend”): operational / strategic discussion / decision / workshop
  • Attendees (and the decision owner, if any)
  • Time box + date/time constraints (and remote/hybrid details)
  • Context links or a 5–10 bullet background summary

Missing-info strategy

  • Ask up to 5 questions from references/INTAKE.md.
  • If answers aren’t available, proceed with explicit assumptions and label unknowns.

Outputs (deliverables)

Produce a Meeting Pack in Markdown (in-chat; or as files if requested), in this order:

  1. Meeting brief (goal, type, decision(s), roles, pre-work)
  2. Pre-read (context, options, open questions; “discover” happens async)
  3. Agenda (timed, discussion-first)
  4. Facilitation script (priming, prompts, how to handle stalls)
  5. Notes template + Decision log + Action items table
  6. Follow-up message (decisions, owners, deadlines, who-else-needs-to-know)
  7. Risks / Open questions / Next steps (always included)

Templates: references/TEMPLATES.md

Workflow (7 steps)

1) Decide if a meeting is needed and pick the format

  • Inputs: Request + context + constraints.
  • Actions: Clarify objective; decide whether the live meeting is for discussion only; pick meeting type (operational vs strategic vs decision vs workshop).
  • Outputs: Meeting type recommendation + meeting objective + draft brief header.
  • Checks: You can answer: “What will be different after this meeting?”

2) Define the discussion scope (what is in-room vs async)

  • Inputs: Objective; stakeholders; open questions.
  • Actions: Separate Discover / Discuss / Decide:
    • Discover (context/data/options) → async pre-read
    • Discuss (trade-offs, objections, clarifications) → live meeting
    • Decide (final call + recording) → after discussion, captured explicitly
  • Outputs: Pre-read outline + list of discussion questions + decision statement(s).
  • Checks: At least 60% of the live time is reserved for discussion, not reading.

3) Prime participants (pre-work + pre-read)

  • Inputs: Pre-read outline; attendee list.
  • Actions: Create a priming section (goal, conversation category, decision rights); assign pre-work; send pre-read 24–48h before when possible.
  • Outputs: Complete pre-read + pre-work assignments.
  • Checks: Every attendee can arrive “ready to discuss” without a live info dump.

4) Build a timed agenda that matches the format

  • Inputs: Meeting type; discussion questions; time box.
  • Actions: Timebox segments; limit to 1–3 topics; include explicit decision points; reserve time for closing alignment.
  • Outputs: Agenda with timing, prompts, and expected outputs per section.
  • Checks: Agenda maps each segment to a tangible output (decision, list, owner, next step).

5) Assign roles and write the facilitation script

  • Inputs: Agenda; participants.
  • Actions: Assign facilitator, decision owner, note-taker, timekeeper; write prompts and “if stuck” branches; set ground rules (parking lot, stack, time boxing).
  • Outputs: Facilitation script + roles list.
  • Checks: Roles are named; escalation path is clear if consensus fails.

6) Run the meeting and capture decisions/actions in real time

  • Inputs: Script; notes template.
  • Actions: Start with priming; facilitate discussion; capture decisions and action items as they happen; keep to time boxes.
  • Outputs: Filled notes + decision log entries + action items table.
  • Checks: No important decision is left implicit or “to be clarified later.”

7) Close, follow up, and improve meeting hygiene

  • Inputs: Captured notes/decisions/actions.
  • Actions: End with: (1) What did we decide? (2) Who does what by when? (3) Who else needs to know? Send follow-up within 24h; propose hygiene changes (split strategic vs operational, cancel/shorten recurring meetings).
  • Outputs: Follow-up message + updated trackers + hygiene recommendations + Risks/Open questions/Next steps.
  • Checks: Every action item has an owner and due date; stakeholders to inform are named.

Quality gate (required)

Examples

Example 1 (Operational cadence): “Design a weekly operational meeting for a product squad to track priority projects without drifting into strategy.”
Expected: a structured meeting brief, recurring agenda template, status update pattern, and a hygiene rule to move strategy to a separate forum.

Example 2 (Decision meeting): “Create a decision meeting pack for choosing between two onboarding flows (A vs B) with PM/Design/Eng and a decision owner.”
Expected: pre-read with options + evidence, timed agenda, facilitation prompts for trade-offs, and a decision record + action items.

Boundary example (redirect): “I need to plan a 2-day strategy offsite with breakout sessions, team dinners, and a follow-up plan.” Response: This is an offsite, not a single meeting. Redirect to running-offsites for the full offsite pack (brief, run-of-show, prework, logistics, post-offsite plan). If one session within the offsite needs a focused meeting pack, handle that session here.

Boundary example (reframe): “Schedule a meeting to ‘get alignment’ but there’s no decision, no owner, and no pre-read.” Response: ask for the decision/outcome; recommend async discovery first; refuse to produce a meeting pack until objective/owner is defined.

Anti-patterns (common failure modes)

  1. ”Status meeting in disguise” -- Labeling a meeting as a “discussion” but filling the agenda with round-robin updates. Live time should be for discussion and decisions; status belongs in async pre-reads.
  2. No pre-read sent -- Spending the first 20 minutes of a 30-minute meeting giving context. Participants arrive cold, discussion time is crushed, and decisions get deferred.
  3. Too many topics -- Cramming 5+ agenda items into a single meeting. Each topic gets shallow treatment and nothing gets decided. Limit to 1-3 topics per meeting.
  4. Missing decision capture -- Good discussion happens but no one writes down what was decided, by whom, or by when. The same discussion repeats next week.
  5. No follow-up within 24h -- Decisions and action items stay in the facilitator’s notes. Stakeholders who were not present never learn the outcomes. Alignment decays.

Reference files

Capabilities

skillsource-liqiongyuskill-running-effective-meetingstopic-agent-skillstopic-ai-agentstopic-automationtopic-claudetopic-codextopic-prompt-engineeringtopic-refoundaitopic-skillpack

Install

Quality

0.47/ 1.00

deterministic score 0.47 from registry signals: · indexed on github topic:agent-skills · 49 github stars · SKILL.md body (8,780 chars)

Provenance

Indexed fromgithub
Enriched2026-04-22 00:56:25Z · deterministic:skill-github:v1 · v1
First seen2026-04-18
Last seen2026-04-22

Agent access