meeting-memo
Skill1 starsChanged 8 months agoMITView on GitHub
Create structured meeting memos from transcription files (.vtt, .txt, .srt) and save them to the second-brain knowledge system. Use when the user asks to create a memo from a meeting, process a meeting transcription, document a check-in or 1:1, or write meeting notes. Triggers on phrases like "create a memo", "meeting memo", "memo from meeting", "process transcription", "meeting notes from", "check-in with". Optionally enriches context by fetching linked Notion project pages. Output is an Obsidian-compatible markdown file with frontmatter, structured sections, action items, and knowledge graph connections.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- skills-agents-co/skills-and-agents-librarytheir own version, first seen Oct 15, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- skills-agents-co/skills-and-agents-library★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository6 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- appunite-brandSkill
- communication-analystSkill
- roam-transcript-syncSkill
- shugyo-mcpSkill
- spec-interviewSkill
The file
We split this skill into the rules it gives an agent
12 rules. About 0 are common, 11 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈11
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 25
Use the Skill tool with structured parameters:
- Distinctive: line 40
When using shorthand, Claude will prompt for any missing required parameters or use defaults.
- Distinctive: line 94
CRITICAL SECURITY STEP: All inputs MUST be sanitized before processing to prevent path traversal, YAML injection, and other attacks.
- Very specific: line 98
Transcription file: Read the .vtt, .srt, or .txt file.
- Distinctive: line 101
Project reference (optional): If the user provides a Notion URL, validate and fetch it:
- Distinctive: line 130
Run steps 1 and 2 in parallel when both inputs are available.
- Distinctive: line 162
When a Notion project page is available:
- Distinctive: line 165
Use the project's deliverable list as a checklist for the status table
- Distinctive: line 169
Follow the format in references/memo-format.md.
- Distinctive: line 193
Preserve strategic tension: Document disagreements faithfully.
- Distinctive: line 194
Action items with owners: Every commitment needs a name and deadline.
- Distinctive: line 196
Polish diacritics: Always use ą, ć, ę, ł, ń, ó, ś, ź, ż when writing Polish.
≈ “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
Meeting Memo
Transform meeting transcriptions into structured, insight-rich memos integrated into the second-brain knowledge system.
Skill Invocation
Use the Skill tool with structured parameters:
/meeting-memo \
--vault-path "/Users/<<user>>/Documents/second-brain" \
--meetings-dir "2-Areas/Meetings" \
--daily-notes-dir "DailyNotes"Or use the shorthand invocation:
Create a memo from this meetingWhen using shorthand, Claude will prompt for any missing required parameters or use defaults.
Workflow
Step 0: Parse and Validate Skill Arguments (MANDATORY)
MANDATORY FIRST STEP: Extract and validate all skill parameters before proceeding:
# Import validation functions
import sys
from pathlib import Path
# Add shared validation module to path
skills_path = Path(__file__).parent.parent
sys.path.insert(0, str(skills_path / "_shared"))
from validation import (
validate_vault_config,
sanitize_participant_name,
sanitize_meeting_title,
sanitize_meeting_metadata,
normalize_unicode,
validate_no_script_mixing,
validate_notion_url,
validate_file_size,
validate_directory_safe,
safe_yaml_frontmatter,
sanitize_wikilink,
atomic_append_to_file,
ValidationError,
SecurityError
)
# Parse skill arguments from invocation
vault_path = Path(args.vault_path).expanduser()
meetings_dir = args.meetings_dir
daily_notes_dir = args.daily_notes_dir
# SECURITY: Validate vault configuration
if not vault_path.exists():
raise ValidationError(f"Vault path does not exist: {vault_path}")
if not vault_path.is_dir():
raise ValidationError(f"Vault path is not a directory: {vault_path}")
# Validate that relative paths don't escape vault
meetings_full = vault_path / meetings_dir
daily_notes_full = vault_path / daily_notes_dir
validate_directory_safe(meetings_full, vault_path)
validate_directory_safe(daily_notes_full, vault_path)CRITICAL SECURITY STEP: All inputs MUST be sanitized before processing to prevent path traversal, YAML injection, and other attacks. NEVER skip this step, even if inputs appear to come from trusted sources. Defense in depth is essential.
Step 1: Gather and Validate Inputs
- Transcription file. Read the
.vtt,.srt, or.txtfile. If the user has a file open in the IDE, use that. Otherwise ask for the path.- SECURITY: Validate file size before reading:
validate_file_size(file_path)
- SECURITY: Validate file size before reading:
- Project reference (optional). If the user provides a Notion URL, validate and fetch it:
# SECURITY: Validate Notion URL before fetching if notion_url: safe_url = validate_notion_url(notion_url) # Then fetch via Notion MCP tool - Meeting folder. Determine and validate the output path using validated skill args:
# SECURITY: Sanitize all path components safe_person = sanitize_participant_name(person_name) safe_date = date # Already in YYYY-MM-DD format safe_title = sanitize_meeting_title(title) # Construct path using validated skill arguments from Step 0 target_dir = vault_path / meetings_dir / safe_person # SECURITY: Validate directory is within vault boundaries validate_directory_safe(target_dir, vault_path) # Create directory target_dir.mkdir(parents=True, exist_ok=True) # Construct filename: YYYY-MM-DD topic.md filename = f"{safe_date} {safe_title}.md" output_path = target_dir / filename
Run steps 1 and 2 in parallel when both inputs are available.
Step 2: Analyze and Sanitize Transcription
Extract from the raw transcription and sanitize all extracted data:
- Participants -- identify speakers from
<v>tags or context# SECURITY: Sanitize participant names raw_participants = extract_participants(transcription) participants = [sanitize_participant_name(p) for p in raw_participants] - Topics discussed -- group by theme, not chronology
- Decisions made -- explicit commitments or conclusions
- Action items -- who committed to what, by when
- Risks and blockers -- anything flagged as a concern
- Strategic discussions -- substantive debates, divergent viewpoints, unresolved questions
- Numbers and specifics -- costs, timelines, percentages, thresholds
- Case studies or examples -- client situations, parallel cases discussed
- Cross-references -- mentions of people, projects, tools, or documents
# SECURITY: Sanitize all wikilinks raw_wikilinks = extract_wikilinks(content) safe_wikilinks = [sanitize_wikilink(link) for link in raw_wikilinks]
Handle bilingual transcriptions (Polish/English mix is common). Transcription quality may be low -- infer meaning from context, do not reproduce garbled text.
Step 3: Enrich with Project Context
When a Notion project page is available:
- Compare discussed status against the project's documented scope and milestones
- Note discrepancies between the project plan and the conversation
- Use the project's deliverable list as a checklist for the status table
Step 4: Write the Memo with Secure Frontmatter
Follow the format in references/memo-format.md.
SECURITY: Generate frontmatter using the safe YAML function:
# Generate secure YAML frontmatter (prevents YAML injection - CVE-2)
frontmatter = safe_yaml_frontmatter(
title=f"{person_name} / {other_person} - {topic}",
description=meeting_description,
person_name=person_name, # Already sanitized
date=date,
project=project_name if project_name else None,
tags=['meeting']
)
# Write memo content
memo_content = frontmatter + "\n\n" + body_contentKey principles:
- Decisions over discussion. Prioritize what was decided and committed.
- Risks and blockers first. Surface anything that could delay progress.
- Quantify. Capture numbers -- costs, timelines, thresholds.
- Preserve strategic tension. Document disagreements faithfully; do not flatten into consensus.
- Action items with owners. Every commitment needs a name and deadline.
- Connect to the knowledge graph. Link to people, projects, organizations via
[[wikilinks]](all sanitized). - Polish diacritics. Always use ą, ć, ę, ł, ń, ó, ś, ź, ż when writing Polish.
Include only sections that have substance. Do not pad with empty sections.
Step 5: Save, Journal Entry, and Report
- Write the memo to the appropriate path in the second-brain:
# Write memo with proper encoding output_path.write_text(memo_content, encoding='utf-8') Add a journal entry to the daily note with file locking to prevent race conditions (CVE-3):
# SECURITY: Use atomic append with file locking # Use validated daily_notes_dir from skill args (Step 0) daily_note_path = vault_path / daily_notes_dir / year / month / f"{date}.md" # Format journal entry journal_entry = f"- [[{person_name}]] -- [[{filename}|{topic}]]\n" journal_entry += " - Key takeaway 1\n" journal_entry += " - Key takeaway 2\n" journal_entry += " - Key takeaway 3\n" # SECURITY: Atomic append prevents race conditions atomic_append_to_file(daily_note_path, journal_entry)The entry must include:
- The meeting link line (person wikilink + memo link + short topic)
- A nested bullet list of key takeaways (3-5 bullets): decisions made, risks surfaced, and critical action items. Keep each bullet to one concise sentence.
- Report a concise summary to the user: key takeaways, number of action items, any risks surfaced.
Tone
Crisp, authoritative, consulting-grade. No filler, no meta-commentary. Professional clarity.
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.

