cortex
Capture conversation summaries and web content to daily notes. Use when user says "save this", "summarize", "capture notes", or pastes a URL without context.
What this skill does
# Cortex
Capture conversation summaries and web content to daily notes synced to GitHub.
## Conversation Summary Mode
When the user wants to save or summarize the current conversation:
### Step 1: Ask for Instructions
Ask: "Any specific focus or instructions for the summary?"
### Step 2: Ask About Topics
Ask: "Any topics to link this to? (e.g., [[AI Coding]], [[Architecture]])"
Before suggesting new topics, search for existing notes that match. If found:
- Single match: "Should I add this to [[AI Coding]]?"
- Multiple matches: "Found related notes: [[AI Coding]], [[AI Tools]]. Which one?"
### Step 3: Extract Key Ideas and Todos
Review the conversation and extract:
- Main concepts and decisions discussed
- **Action items** - these go in the `# Todos` section at the top of the daily note
- Conclusions and decisions made
- Any URLs or resources mentioned
### Step 4: Create/Update Topic Docs (if specified)
If user wants to link to a topic:
- Fetch existing topic file (or create new)
- Add a reference to today's daily note with 1-sentence summary
- Push updated topic file
### Step 5: Format as Daily Note Entry
Create an entry with timestamp, key ideas, related notes, and links.
**Timestamp:** Read `timezone` from `scripts/config.json` and calculate the current local time. Use format `HH:MM` (24-hour).
### Step 6: Save to GitHub
Fetch existing daily note, append new entry, push back.
### Example Conversation
```
User: Save this conversation
Claude: Any specific focus or instructions for the summary?
User: Focus on the architecture decisions we made
Claude: Any topics to link this to? I found [[Event-Driven Architecture]] in your notes.
User: Yes, add it there. Also create a new note for System Design.
Claude: Got it. I'll update [[Event-Driven Architecture]] and create [[System Design]].
Added to Todos section:
- [ ] Implement retry logic with exponential backoff
- [ ] Document event schema for notification system
Added to Cortex section (2026-01-03.md):
## 14:30 - Architecture Discussion
### Key Ideas
- Decided to use event-driven architecture for the notification system
- Will implement retry logic with exponential backoff
- See [[Event-Driven Architecture]] for more details
### Related Notes
- [[Event-Driven Architecture]]
- [[System Design]]
### Links
- [EDA Article](https://example.com/eda) - reference article discussed
Updated [[Event-Driven Architecture]] with link to today's discussion.
Created [[System Design]] with initial notes.
```
## Link Summary Mode
When the user pastes a URL without context:
### Step 1: Offer to Summarize
Ask: "Would you like me to summarize this link?"
### Step 2: Get Focus Instructions
If yes, ask: "Any specific focus or instructions for the summary?"
### Step 3: Fetch and Summarize
Retrieve the page content and create a summary based on instructions.
### Step 4: Save to Daily Notes
Add the link with its summary to the daily notes.
### Example Conversation
```
User: https://example.com/article-about-microservices
Claude: Would you like me to summarize this link?
User: Yes, focus on the deployment strategies
Claude: Any topics to link this to? I found [[Microservices]] in your notes.
User: Yes, and also [[Deployment Strategies]]
Claude: Got it. Keep it brief?
User: Yes
[Fetches and summarizes the article]
Added to Todos section:
- [ ] Set up blue-green deployment for checkout service
- [ ] Evaluate feature flag tooling options
Added to Cortex section (2026-01-03.md):
## 15:00 - Link Summary
### Key Ideas
- Article recommends blue-green deployments for microservices
- Canary releases reduce risk for critical services
- Feature flags enable gradual rollouts
### Related Notes
- [[Microservices]]
- [[Deployment Strategies]]
### Links
- [Microservices Deployment Strategies](https://example.com/article-about-microservices) - deployment patterns
Updated [[Microservices]] with link to this article.
```
## Daily Notes Format
Files are named `YYYY-MM-DD.md` (one file per day). Multiple sessions throughout the day all append to the same file.
### File Structure
Daily notes have two main sections:
1. **`# Todos`** - Action items at the top of the file
2. **`# Cortex`** - Timestamped entries at the bottom
### Ordering Multiple Entries
When appending a new entry:
1. Fetch existing daily note
2. Extract any action items/todos from the conversation
3. Add todos to the `# Todos` section (create at top if missing)
4. Find the `# Cortex` section (or create it below todos if missing)
5. Parse existing entries by their `## HH:MM` timestamps
6. Insert new entry in chronological order (don't intermix - each summary stays as a complete block)
7. Push updated file
The exact timestamp isn't critical - entries just need to be roughly in order of when they happened.
### Todo Extraction
When summarizing a conversation, look for:
- Explicit action items ("I need to...", "TODO:", "Action item:")
- Decisions that require follow-up
- Tasks mentioned but not completed
- Next steps discussed
Format todos as Obsidian tasks:
- `- [ ]` for incomplete tasks
- `- [x]` for completed tasks
Example: `2026-01-03.md`
```markdown
# Todos
- [ ] Implement retry logic with exponential backoff
- [ ] Review EDA article and summarize key patterns
- [ ] Set up blue-green deployment pipeline
# Cortex
## 14:30 - Architecture Discussion
### Key Ideas
- Decided to use event-driven architecture
- Will implement retry logic with exponential backoff
- See [[Event-Driven Architecture]] for more details
### Related Notes
- [[Event-Driven Architecture]]
- [[System Design]]
### Links
- [EDA Article](https://example.com/eda) - reference article
---
## 15:00 - Link Summary
### Key Ideas
- Blue-green deployments recommended for microservices
### Related Notes
- [[Microservices]]
- [[Deployment Strategies]]
### Links
- [Microservices Guide](https://example.com/guide) - deployment patterns
---
```
The `# Todos` section stays at the top for easy access. The `# Cortex` header marks the section for timestamped entries.
### Format Rules
- **Filename**: `YYYY-MM-DD.md`
- **Todos section**: `# Todos` at the top of the file (action items as `- [ ]` tasks)
- **Cortex section**: `# Cortex` below todos
- **Entry header**: `## HH:MM - Title`
- **Key Ideas section**: `### Key Ideas` with bullet points
- **Related Notes section** (optional): `### Related Notes` with `[[Topic Name]]` links - only include if there are related notes (e.g., `[[Claude Code]]`, `[[Architecture Patterns]]`)
- **Links section**: `### Links` with `[Title](URL) - description`
- **Entry separator**: `---` between entries
## Internal Links
Use Obsidian-style `[[wiki links]]` to connect notes.
### Inline Links (wiki-style)
Reference topics naturally in sentences:
```markdown
This discussion covered [[AI Coding]] best practices and [[Architecture Patterns]].
```
### Related Notes Section
List related topics at the end of an entry:
```markdown
### Related Notes
- [[AI Coding]]
- [[Cloudflare]]
- [[Architecture Patterns]]
```
## Topic Documents
Topic docs are standalone notes for recurring themes, projects, or reference material. They live in the same `notes/` folder as daily notes.
### Finding Existing Topics
Before creating a new topic, search the existing notes in the repo:
1. Check for notes with matching or similar names to the topic being discussed
2. If found, suggest: "Should I add this to [[AI Coding]]?"
3. If multiple matches: "Found related notes: [[AI Coding]], [[AI Tools]], [[Coding Best Practices]]. Which one?"
To check if a topic exists, try fetching it:
```bash
cd /mnt/skills/user/cortex/scripts
uv run github_sync.py --fetch "notes/AI Coding.md"
```
### When to Create Topics
- Not everything needs a topic file - use judgment
- Create topics for recurring themes, projects, or reference material
- Daily notes can reference topics for deeper detail
### Topic Document Structure
```markdown
# AI Coding
## Overview
Brief description of the topic.
## NRelated in Writing & Docs
jax-development
IncludedUse this skill when the user is writing, debugging, profiling, refactoring, reviewing, benchmarking, parallelising, exporting, or explaining JAX code, or when they mention JAX, jax.numpy, jit, grad, value_and_grad, vmap, scan, lax, random keys, pytrees, jax.Array, sharding, Mesh, PartitionSpec, NamedSharding, pmap, shard_map, Pallas, XLA, StableHLO, checkify, profiler, or the JAX repo. It helps turn NumPy or PyTorch-style code into pure functional JAX, fix tracer/control-flow/shape/PRNG bugs, remove recompiles and host-device syncs, choose transforms and sharding strategies, inspect jaxpr/lowering/IR, and benchmark compiled code correctly.
nature-article-writer
IncludedDrafts, rewrites, diagnostically critiques, and style-calibrates primary research manuscripts for Nature and Nature Portfolio journals. Use when the user wants a Nature-style title, summary paragraph or abstract, introduction, results, discussion, methods, figure legends, presubmission enquiry, cover letter, reviewer response, or when a scientific draft sounds generic, jargon-heavy, structurally weak, or AI-ish and needs precise, broad-reader-friendly prose without inventing data, analyses, or references. Best for primary research articles and letters rather than reviews or press releases unless explicitly adapting one.
deckrd
IncludedDocument-driven framework that derives requirements, specifications, implementation plans, and executable tasks from goals through structured AI dialogue. Use when user says "write requirements", "create spec", "plan implementation", "derive tasks", "structure this feature", "break down into tasks", or "document this module". Also use for reverse engineering existing code into docs (/deckrd rev). Do NOT use for direct code writing — use /deckrd-coder after tasks are generated. Do NOT use when the user only wants to run or fix existing code without planning.
clinical-decision-support
IncludedGenerate professional clinical decision support (CDS) documents for pharmaceutical and clinical research settings, including patient cohort analyses (biomarker-stratified with outcomes) and treatment recommendation reports (evidence-based guidelines with decision algorithms). Supports GRADE evidence grading, statistical analysis (hazard ratios, survival curves, waterfall plots), biomarker integration, and regulatory compliance. Outputs publication-ready LaTeX/PDF format optimized for drug development, clinical research, and evidence synthesis.
handling-sf-data
IncludedSalesforce data operations with 130-point scoring. Use this skill to create, update, delete, bulk import/export, generate test data, and clean up org records using sf CLI and anonymous Apex. TRIGGER when: user creates test data, performs bulk import/export, uses sf data CLI commands, needs data factory patterns for Apex tests, or needs to seed/clean records in a Salesforce org. DO NOT TRIGGER when: SOQL query writing only (use querying-soql), Apex test execution (use running-apex-tests), or metadata deployment (use deploying-metadata).
accelint-ac-to-playwright
IncludedConvert and validate acceptance criteria for Playwright test automation. Use when user asks to (1) review/evaluate/check if AC are ready for automation, (2) assess if AC can be converted as-is, (3) validate AC quality for Playwright, (4) turn AC into tests, (5) generate tests from acceptance criteria, (6) convert .md bullets or .feature Gherkin files to Playwright specs, (7) create test automation from requirements. Handles both bullet-style markdown and Gherkin syntax with JSON test plan generation and validation.