limitless-analysis
Deep review of the day's Limitless recordings -- catch things meetings missed, extract insights, and prep tomorrow's context.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Deep review of the day's Limitless recordings -- catch things meetings missed, extract insights, and prep tomorrow's context.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | limitless-analysis |
| description | Deep review of the day's Limitless recordings -- catch things meetings missed, extract insights, and prep tomorrow's context. |
Deep analysis of today's Limitless lifelog recordings. The user wears a Limitless AI pendant that captures all conversations throughout the day -- both scheduled meetings and impromptu conversations. This skill reviews everything that was captured and surfaces what matters.
Meetings captured by Fireflies have structured transcripts. But a lot of important information flows through conversations that aren't formal meetings -- phone calls, hallway chats, thinking-out-loud moments. Limitless captures all of it. This skill reviews the full day and extracts what Edwin needs to know.
Read all files matching: data/limitless/lifelogs/YYYY-MM/YYYY-MM-DD-*.md
Each file is a conversation segment with:
Read today's calendar (data/o365/calendar/YYYY-MM/YYYY-MM-DD.md and data/google/calendar/YYYY-MM/YYYY-MM-DD.md).
For each Limitless segment, determine:
data/calls/ for matching timeframe)Read data/fireflies/transcripts/YYYY-MM/YYYY-MM-DD-*.md.
Identify which Limitless segments are already covered by Fireflies (better quality transcripts). Focus Limitless analysis on segments NOT covered by Fireflies.
For each Limitless segment that isn't already covered by Fireflies:
For each non-Fireflies segment:
Across all segments:
Write analysis to: ~/Edwin/briefing-book/docs/5. Overnight/logs/YYYY-MM-DD-limitless.md
Format:
---
date: YYYY-MM-DD
type: limitless-analysis
segments: [count]
off_calendar: [count]
---
# Limitless Day Review -- YYYY-MM-DD
## Day Overview
[2-3 sentences: what kind of day was it? Heavy meetings? Lots of phone calls? Quiet heads-down?]
## Off-Calendar Conversations
[These are the gold -- conversations Fireflies didn't capture]
### [Time] -- [Brief description]
- **Who:** [participants if identifiable]
- **Key points:** [what was discussed]
- **Action items:** [if any]
- **Context:** [why this matters]
## Emotional Arc
[How was the day from an energy/stress perspective? What moments were high/low?]
## People Map
[Who did the user interact with today and in what context?]
## Tomorrow Context
[What from today should inform tomorrow's brief or Edwin's behavior?]
## PM Items Extracted
[List any new PM items added, or note if none]
After writing, publish to Obsidian:
python3 briefing-book/scripts/obsidian-publish --all
SKILL_COMPLETE: limitless-analysis
STATUS: success | error
ARTIFACT: ~/Edwin/briefing-book/docs/5. Overnight/logs/YYYY-MM-DD-limitless.md
SEGMENTS_REVIEWED: [count]
OFF_CALENDAR: [count]
PM_ITEMS_EXTRACTED: [count]
ERRORS: [any errors, or "none"]
Meeting-to-closure decision tracking -- harvest decisions actually made in the last week's meetings (the weekly priorities forum first), maintain a persistent ledger, verify downstream follow-through evidence, and surface quietly-dying decisions at the two-week mark.
Monthly red-team pass over the user's active big bets -- build the strongest evidence-backed case AGAINST each position they're invested in (steelman opposition, not strawman compliance), ending each bet with an honest KILL / HEDGE / PROCEED-EYES-OPEN verdict.
Weekly knowledge-graph curation loop -- freshness sweep against the week's reality, stale-edge invalidation, pending-merge queue drain, and org-chart temporal completeness against your authority model.
PM Weekly Triage
Decision radar + dossier generator -- detect decisions approaching the user in the next few days and build evidence-backed dossiers (prior positions, stakeholder map, constraining commitments, what it gates) before the moment arrives.
Draft-first email posture -- high-priority emails and key-contact unreplied threads arrive each morning WITH a reply already drafted in the user's voice. The user edits-and-sends or discards instead of starting cold. Drafts only, never sends.