Agent skill
marta-meeting
Install this agent skill to your Project
npx add-skill https://github.com/ChinchillaEnterprises/ChillSkills/tree/main/marta-meeting
SKILL.md
Marta Meeting Processing
Process post-meeting artifacts from Marta/BeonIQ meetings. Fetches transcripts, saves notes, updates context, syncs GitHub issues, and posts summaries to Slack.
Trigger: Use when user says "process the meeting", "grab the marta meeting", "meeting notes", or after any meeting with Marta/BeonIQ team.
Workflow
Phase 1: Fetch Transcript
-
Fathom API (primary source):
bash# List recent calls curl -s -H "X-Api-Key: 2Oio2EqggEIt5g_XuDt3KA.8ubzimd9viyHFQw-kc4A-uXfuDCS7kKOwC3bUGY5Np8" \ "https://api.fathom.ai/external/v1/calls?limit=10" | jq '.calls[:5] | .[] | {id, title, started_at}'Then fetch the specific call:
bashcurl -s -H "X-Api-Key: 2Oio2EqggEIt5g_XuDt3KA.8ubzimd9viyHFQw-kc4A-uXfuDCS7kKOwC3bUGY5Np8" \ "https://api.fathom.ai/external/v1/calls/<CALL_ID>" | jq '.' -
Granola (backup):
- Cache at
/Users/tori/Library/Application Support/Granola/cache-v4.json - Contains metadata + transcripts (notes/summaries are server-side only)
- Cache at
Phase 2: Save Meeting Notes
Save to: context/team-communications/YYYY-MM-DD/meetings/
Filename format: fathom-<descriptive-slug>-<recording-id>.md
Template:
# [Meeting Title]
**Date:** [Date, Time CT]
**Source:** Fathom (Recording ID: [ID])
**Participants:** [Names and roles]
**Artifacts:** [Any shared docs, boards, etc.]
---
## Key Decisions
### 1. [Decision Title] (CONFIRMED/PROPOSED/DISCUSSED)
- Detail
- Quote if impactful
---
## Action Items
| Owner | Action | Priority |
|-------|--------|----------|
| Name | What to do | High/Medium/Low |
---
## Architecture Impact on Existing Issues
- **#NNN** (issue title): How this decision affects the issue
---
## GitHub Issues Created
| Issue | Title | Assignees | Labels |
|-------|-------|-----------|--------|
| #NNN | Title | @user | label1, label2 |
*(Populated during Phase 4c)*
---
*Captured from Fathom transcript. [Any availability notes]*
Phase 3: Update Context File
Update context/marta-mvp.md with:
- New decisions that change architecture or domain model
- New action items
- Updated subscription/pricing info
- New UX decisions or deferred topics
- New meeting notes file reference
Phase 4: GitHub Issues
Phase 4a: Comment on Existing Issues
For each issue affected by meeting decisions:
- Add a comment with the relevant decision
- Quote the speaker if it clarifies intent
- Reference the meeting notes file
- Update labels if status changed
Comment template:
## Update from [Date] Meeting
[Decision or finding]
> "[Quote from participant]" — [Name]
See: `context/team-communications/YYYY-MM-DD/meetings/[filename].md`
Phase 4b: Dedup Check
Before creating ANY new issue, search for duplicates:
gh issue list --repo ChinchillaEnterprises/transportation-insight --search "key words from action item" --state open --json number,title,assignees,labels
If >70% scope overlap with an existing open issue: Comment on the existing issue instead of creating a new one. Add the new context/requirements from the meeting.
Phase 4c: Auto-Create New Issues
For each new action item with clear scope, create an issue using the assignment and label rules below.
gh issue create \
--repo ChinchillaEnterprises/transportation-insight \
--title "[PREFIX] Issue title" \
--assignee user1,user2 \
--label "label1,label2" \
--body "$(cat <<'EOF'
## Source
[Date] meeting with [participants]
## Context
[Why this issue exists, what meeting decision drove it]
## Requirements
- [ ] Requirement 1
- [ ] Requirement 2
## Relationship to Existing Issues
- Depends on #NNN
- Blocks #NNN
- Related to #NNN
## Acceptance Criteria
- [ ] Criterion 1
- [ ] Criterion 2
EOF
)"
After each issue is created, record it in the meeting notes ## GitHub Issues Created table.
Phase 4d: Post-Creation Tracking
After all issues are created:
- Update the meeting notes file with the
## GitHub Issues Createdtable - Include created issue numbers in the Slack summary (Phase 5)
Team Assignment Rules
Map action items to assignees based on domain keywords in the action item content.
| GitHub Username | Person | Domains |
|---|---|---|
feathars91 |
Abel | backend, Lambda, data, pipeline, Databricks, architecture, DevOps, AWS, API, SQL, EDW, metrics, anomaly detection, deployment, server, database, schema, migration, ETL, S3, CloudWatch, CDK, SST, config, environment |
SOROUSH911 |
Saroush | frontend, UI, component, chart, styling, MUI, layout, dashboard, modal, sidebar, visualization, responsive, CSS, theme, color, icon, table, grid, card, widget, tooltip, animation |
hailee-landin |
Hailee | QA, testing, validation, regression, verification, data accuracy, test plan, smoke test, UAT |
jamie805 |
Jamie | mockup, coordination (only if explicitly assigned to Jamie in meeting) |
Multi-assignee rules:
- Full-stack features (touches both UI + backend) →
feathars91,SOROUSH911 - Features with explicit QA acceptance criteria or "Hailee to verify" → add
hailee-landin - Ambiguous domain or purely technical → default
feathars91 - Never assign
mbras(Marta is the client, not a developer)
How to apply: Read the action item content. Match keywords against the Domains column. If multiple people's domains match, assign all of them. If nothing matches, default to feathars91.
Label Assignment Rules
Apply labels in three categories. Use existing repo labels only.
1. Type (exactly one)
| Label | When to use |
|---|---|
feature |
New functionality |
bug |
Something broken |
architecture |
System design, schema, infrastructure |
Data |
Data pipeline, EDW, metrics, SQL |
EPIC |
Large multi-issue initiative |
maintenance |
Cleanup, refactoring, tech debt |
documentation |
Docs, README, wiki |
2. Priority (if discussed in meeting)
| Label | When to use |
|---|---|
Critical |
Must ship this sprint, blocker |
High |
Important, next up |
Medium |
Should do soon |
3. Context (all that apply)
| Label | When to use |
|---|---|
MVP |
Required for MVP launch |
Blocked |
Cannot start yet (add reason in body) |
blocked:external |
Waiting on client/vendor |
blocked:data |
Waiting on data access or pipeline |
blocked:agent |
Waiting on AI agent capability |
Issue Title Prefix Rules
Every issue title MUST start with a prefix in brackets:
| Prefix | When to use |
|---|---|
[FEATURE] |
New user-facing functionality |
[DATA] |
Data pipeline, EDW, metrics, SQL changes |
[PLANNING] |
Scoping, research, design decisions |
[UI] |
Frontend-only visual/layout changes |
[BUG] |
Defect fix |
[PERF] |
Performance optimization |
[HANDOFF] |
Cross-team handoff or integration point |
[QA] |
Test plan, validation task |
[ARCH] |
Architecture, infrastructure, schema |
[OPS] |
DevOps, deployment, CI/CD |
[EPIC] |
Epic tracking issue |
Phase 5: Post to Slack
Post summary to #marta channel (C09QU6TN2EQ), tagging Marta (<@U095VLYG3C7>):
Keep it SHORT — 3-5 bullets max. Link to the meeting notes file or GitHub issues for details. Include any newly created issue numbers.
Also post to #internal-beoniq (C09JD94HRUN) with:
- Key architectural decisions
- Action items for Abel/dev team
- Any blockers or priority changes
- List of created/updated GitHub issues with assignees
Key Contacts
| Person | Slack ID | GitHub | Role |
|---|---|---|---|
| Marta (mbras) | U095VLYG3C7 |
mbras |
Client, UX approver (NEVER assign issues) |
| Abel | U072PCVTS00 |
feathars91 |
Backend dev |
| Saroush | — | SOROUSH911 |
Frontend dev |
| Hailee | — | hailee-landin |
QA + issue logging |
| Jamie | — | jamie805 |
Project lead (CLI user) |
File Locations
| What | Where |
|---|---|
| Meeting notes | context/team-communications/YYYY-MM-DD/meetings/ |
| Context file | context/marta-mvp.md |
| This skill | ~/.claude/skills/marta-meeting/SKILL.md |
| Fathom API key | In this file (above) |
| Granola cache | /Users/tori/Library/Application Support/Granola/cache-v4.json |
Checklist
[ ] Fetch transcript (Fathom → Granola fallback)
[ ] Save meeting notes to team-communications (include GitHub Issues Created table)
[ ] Update marta-mvp.md context
[ ] Comment on existing GitHub issues affected by decisions (4a)
[ ] Dedup check — search before creating any new issue (4b)
[ ] Auto-create new issues with assignees + labels (4c)
[ ] Record created issues in meeting notes file (4d)
[ ] Post summary to #marta (include new issue numbers)
[ ] Post dev priorities to #internal-beoniq (include assignees + issue numbers)
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
Red Team
Use this skill when the user says "red team", "stress test", "attack this", "critique this", "run the models against this", "test this plan", "refine this with AI", or wants to use external AI models to critique, validate, or improve a document, pitch, plan, or idea. Also use when the user wants to save Claude Code usage by offloading analysis to other models.
Upwork Scanner
This skill should be used when the user asks to "scan upwork", "find upwork jobs", "check upwork", "run the scanner", "look for jobs", or mentions finding freelance projects for Chinchilla AI.
Polymarket Arb Scanner
This skill should be used when the user asks to "scan polymarket", "find arb opportunities", "check polymarket arbs", "run arb scanner", "polymarket arbitrage", or mentions detecting price inefficiencies on Polymarket.
captain-update
Polymarket Copy Trader
This skill should be used when the user asks to "copy trade", "telegram bot polymarket", "follow traders", "copy polymarket", or mentions building a copy trading bot for Polymarket.
Architecture Diagram
This skill should be used when the user asks to "generate a diagram", "create an architecture diagram", "make a diagram", "draw a system diagram", or needs a branded Chinchilla AI technical diagram for proposals or documentation.
Didn't find tool you were looking for?