foundation-meeting-recap
Skill777 starsChanged 10 months agoApache-2.0View on GitHub
Produces a topic-segmented post-meeting summary for attendees with decisions highlighted and actions captured inline per topic (plus a consolidated action view at the end). Auto-populates topic skeleton from a sibling meeting-agenda when available and reconciles planned vs. actual topics. Accepts transcripts from Zoom, Meet, Otter, Fireflies, Krisp MCP, or manual notes; runs on variable-quality input without blocking. For synthesizing user research interviews across participants, use discover-interview-synthesis.
In this repository69 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- define-hypothesisSkill
- define-jtbd-canvasSkill
- define-opportunity-treeSkill
- define-prioritization-frameworkSkill
- define-problem-statementSkill
- deliver-acceptance-criteriaSkill
- deliver-edge-casesSkill
- deliver-launch-checklistSkill
- deliver-prdSkill
- deliver-release-notesSkill
- deliver-user-storiesSkill
- develop-adrSkill
- develop-design-rationaleSkill
- develop-solution-briefSkill
- develop-spike-summarySkill
- discover-competitive-analysisSkill
- discover-interview-synthesisSkill
- discover-journey-mapSkill
- discover-market-sizingSkill
- discover-stakeholder-summarySkill
- foundation-build-risk-reviewSkill
- foundation-lean-canvasSkill
- foundation-meeting-agendaSkill
- foundation-meeting-briefSkill
- foundation-meeting-synthesizeSkill
- foundation-okr-writerSkill
- foundation-personaSkill
- foundation-prioritized-action-planSkill
- foundation-stakeholder-briefingsSkill
- foundation-stakeholder-updateSkill
- iterate-lessons-logSkill
- iterate-pivot-decisionSkill
- iterate-refinement-notesSkill
- iterate-retrospectiveSkill
- measure-dashboard-requirementsSkill
- measure-experiment-designSkill
- measure-experiment-resultsSkill
- measure-instrumentation-specSkill
- measure-okr-graderSkill
- measure-survey-analysisSkill
- tool-design-sprint-briefSkill
- tool-design-sprint-decide-and-storyboardSkill
- tool-design-sprint-map-and-targetSkill
- tool-design-sprint-prototype-planSkill
- tool-design-sprint-readinessSkill
- tool-design-sprint-sketchSkill
- tool-design-sprint-test-and-scoreSkill
- tool-foundation-sprint-approach-optionsSkill
- tool-foundation-sprint-basicsSkill
- tool-foundation-sprint-briefSkill
- tool-foundation-sprint-differentiationSkill
- tool-foundation-sprint-founding-hypothesisSkill
- tool-foundation-sprint-magic-lensesSkill
- tool-foundation-sprint-readinessSkill
- tool-note-and-voteSkill
- utility-mermaid-diagramsSkill
- utility-pm-changelog-curatorSkill
- utility-pm-criticSkill
- The first 60 shown.
The file
Show its rules and text
We split this skill into the rules it gives an agent
14 rules. About 0 are common, 9 are distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
Distinctive · estimate
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 24
After any internal meeting that produces decisions or actions affecting attendees
- Very specific: line 25
When a sibling foundation-meeting-agenda exists and needs reconciliation (planned vs.
- Distinctive: line 26
When the team needs a topic-organized reference rather than a chronological transcript dump
- Distinctive: line 27
When inputs include a transcript (Zoom, Meet, Otter, Fireflies, Krisp MCP) or mixed notes and transcript
- Distinctive: line 42
Read all provided inputs (transcript, notes, or hybrid) and note input quality upfront
- Distinctive: line 44
Run inference: meeting metadata from content, decisions from language markers, actions from imperative-future patterns, owners from …
- Very specific: line 48
If invoked with --go, skip the inference summary.
- Very specific: line 56
When the ratio of unassigned actions to total actions exceeds 0.3 (30%), OR when any high-priority action lacks an owner, the skill …
- Distinctive: line 70
When asked to create a meeting recap, follow these steps:
- Very specific: line 76
Look in the same directory for a file matching the pattern {YYYY-MM-DD}_{HH-MMtimezone}_{title-slug}_agenda.md.
- Distinctive: line 82
If agenda was loaded, use its topic list as scaffold (plus any emergent topics)
- Distinctive: line 83
If no agenda, identify topics from transcript discourse markers ("moving on to", "the other thing")
- Distinctive: line 106
When we reconvene, what needs to happen on the critical path before that.
- Very specific: line 109
Every action's owner appears in attendees or is explicitly unassigned
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
<!-- PM-Skills | https://github.com/product-on-purpose/pm-skills | Apache 2.0 -->Meeting Recap
A meeting recap is a post-meeting topic-segmented summary produced for attendees and light distribution. It organizes content by topic rather than chronology, highlights decisions visually, and captures actions inline (with owner, due date, dependencies) per topic segment, plus a consolidated actions view at the end for scannability.
This skill absorbs what would otherwise be a separate "meeting-actions" skill. Actions in this family live alongside the context that makes them meaningful, not in a sibling artifact.
This skill belongs to the Meeting Skills Family. It conforms to the Meeting Skills Family Contract.
When to Use
- After any internal meeting that produces decisions or actions affecting attendees
- When a sibling
foundation-meeting-agendaexists and needs reconciliation (planned vs. actual topics) - When the team needs a topic-organized reference rather than a chronological transcript dump
- When inputs include a transcript (Zoom, Meet, Otter, Fireflies, Krisp MCP) or mixed notes and transcript
When NOT to Use
- Setting the pre-meeting structure attendees will see (topics, owners, time boxes) before the meeting happens. Use
foundation-meeting-agenda. recap summarizes what happened after a meeting; agenda sets expectations before it. - Preparing the user's private strategic positioning before a meeting that matters. Use
foundation-meeting-brief. recap is a shared post-meeting record; brief is pre-meeting tactical prep never shown to attendees. - Communicating outcomes to non-attendees. Use
foundation-stakeholder-update. recap assumes reader context; stakeholder-update translates to readers without it. - Cross-meeting synthesis (patterns across multiple meetings). Use
foundation-meeting-synthesize. - Synthesizing user research interviews across participants, not one meeting's attendees. Use
discover-interview-synthesis. recap organizes a single meeting by topic; interview-synthesis finds patterns across many research conversations. - Live meeting note-taking. This skill consumes finished inputs; it does not transcribe live.
Zero-friction execution
Per the family contract, this skill never blocks on interrogation. Default flow:
- Read all provided inputs (transcript, notes, or hybrid) and note input quality upfront
- Auto-discover related agenda via filename-prefix match on same-directory
*_{title}_agenda.md - Run inference: meeting metadata from content, decisions from language markers, actions from imperative-future patterns, owners from attendee context
- Present a brief inference summary and accept one-word
goor corrections - Produce the recap
If invoked with --go, skip the inference summary. If the user provides all metadata upfront, no checkpoint appears.
Fabrication prohibition
This skill never fabricates owners, decisions, or actions. When an action lacks an explicit owner, it is captured as [owner: unassigned, needs confirmation] not invented. When a decision is implicit ("it sounded like we decided X"), it is flagged with a confidence marker rather than stated as fact. Trust decay from fabrication is worse than the mild friction of flagging.
Ownership reconciliation threshold (v1.1.0)
When the ratio of unassigned actions to total actions exceeds 0.3 (30%), OR when any high-priority action lacks an owner, the skill surfaces a dedicated ## ⚠ Ownership reconciliation required section at the top of the recap (above the topic segments) listing:
- All unassigned actions
- A suggested next step for each (who should probably own this based on topic context, flagged as inference)
- A recommended follow-up action (Slack thread, 15-min sync, async survey)
The shareable summary also leads with this flag when triggered: ⚠ Ownership reconciliation required: N of M actions lack owners.
The unassigned_action_ratio frontmatter field (float 0.0-1.0) records the ratio for downstream tools.
Rationale: a recap with 60% ownerless actions is "non-fabricated" (per the prohibition above) but operationally broken. a pile of broken tickets. The threshold makes this visible instead of silently shipping.
Instructions
When asked to create a meeting recap, follow these steps:
- Parse inputs and detect type Transcript (timestamped speaker-attributed lines), notes (bullet or prose), or hybrid. Note input quality upfront. transcript plus structured notes is high; scrappy bullets is low.
- Auto-discover related agenda
Look in the same directory for a file matching the pattern
{YYYY-MM-DD}_{HH-MMtimezone}_{title-slug}_agenda.md. If found, load it. its topic list is the recap's topic skeleton and itsdesired_outcomesdrive the meeting-quality reconciliation. - Present go-mode inference summary
Show detected meeting date, title, attendees (if inferred), input quality assessment. Accept
goor corrections. - Topic-segment the content
- If agenda was loaded, use its topic list as scaffold (plus any emergent topics)
- If no agenda, identify topics from transcript discourse markers ("moving on to", "the other thing")
- Per topic segment, extract
- Discussion summary: 2-3 sentences capturing what was discussed
- Decisions made: bold-flagged visually. Never fabricate. if uncertain, flag "appears to have decided X [confidence: medium]"
- Actions: owner + due date + dependencies. Flag missing owners as
[owner: unassigned]and missing dates as[due: not specified]. Never invent. - Open questions: unresolved items with confidence marker on whether they are truly unresolved or simply not re-raised
- Consolidate actions by owner Regroup all actions under each owner. Enables single-owner scan ("what do I owe after this meeting?").
- Reconcile agenda (if agenda loaded)
topics_planned: from agenda topic listtopics_hit: topics actually discussedtopics_skipped: planned but not discussed, with brief reasontopics_emerged: discussed but not on agenda
- Assess meeting quality signals
outcomes_achieved: N/M ratio of agendadesired_outcomesmet (when agenda present)started_on_time/ended_on_time: from timestamps when available; skipped when notkey_attendees_present: flag if decision-makers were absent
- Surface next steps When we reconvene, what needs to happen on the critical path before that.
- Render TEMPLATE.md and validate
- Every action's owner appears in
attendeesor is explicitlyunassigned meeting_quality.outcomes_achievedmatchesN/Mpattern when populatedagenda_reconciliationfields present even when empty lists
- Every action's owner appears in
Project Memory Contract
Active only when .claude/pm-skills.local.md exists. With no file, ignore this section entirely
and behave exactly as described above.
- Reads:
active_initiative, so decisions are recorded against the initiative they belong to. - Writes: each decision actually reached in the meeting to the
## Decisionssection, dated and attributed, plus the recap itself as adecisionartifact. - Posture: propose the entry and wait for confirmation before writing, unless
memory_auto_append: trueis set, in which case append and echo what was written. - Write discipline: re-read the file immediately before writing, never from the copy that produced the proposal. If it changed in between, merge your entry into the current state and re-propose rather than overwriting; add only your own entry and leave every other field and section byte-identical. Nothing enforces this at runtime and the file is gitignored, so a careless whole-file write loses another session's work with no way to recover it.
The fabrication prohibition above governs this write without exception: record only decisions that were stated. An unrecorded decision is recoverable; an invented one in durable memory is not. This complements the family's filename-based chaining rather than replacing it: filenames still locate the sibling artifacts of one meeting, while project memory carries the durable product context across meetings.
Quality Checklist
- Input quality flagged honestly (high / medium / low)
- Agenda auto-discovery attempted; result noted in Generation context
- Every topic segment has Discussion / Decisions / Actions / Open questions subsections
- Decisions are bold-flagged; never invented; confidence marked when inferred
- Every action has owner (or explicit
unassigned) + due date (or explicitnot specified) - Consolidated actions view regrouped by owner
- Agenda reconciliation populated when agenda was loaded
- Meeting quality signals populated with data available; skipped fields flagged
- Shareable summary 5-6 lines, lead with decisions and top actions
- Sources and References Generation context notes transcript source and any missing owners/dates
See also
- Meeting Skills Family Contract
foundation-meeting-agenda. upstream: provides topic skeleton and desired-outcomesfoundation-meeting-synthesize. downstream: consumes recaps for cross-meeting synthesisfoundation-stakeholder-update. downstream: translates recap outcomes to non-attendees
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.

