with one click
save
Quick-save key findings, decisions, or notes to the sayou workspace
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Menu
Quick-save key findings, decisions, or notes to the sayou workspace
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | save |
| description | Quick-save key findings, decisions, or notes to the sayou workspace |
| user_invocable | true |
| triggers | ["save this","remember this","save findings","save decision","save note","save what we learned"] |
Save key knowledge from the current conversation to the sayou workspace as a structured, versioned file.
Analyze the conversation — identify the most important knowledge to save:
Choose a logical path based on content type:
decisions/ — architectural decisions, technology choices, design rationaleresearch/ — competitive analysis, market research, technical investigationnotes/ — meeting notes, brainstorms, general notesbugs/ — bug investigations, root causes, fixes appliedguides/ — how-to guides, setup instructions, runbooksWrite with YAML frontmatter — always include structured metadata:
---
type: decision | research | note | bug-report | guide
status: draft | reviewed | approved | archived
topic: short-topic-name
date: YYYY-MM-DD
tags: [relevant, tags]
---
Use workspace_write to save the file. Use a descriptive filename in kebab-case (e.g., auth-token-refresh-fix.md).
Confirm what was saved — show the path, key frontmatter fields, and mention it's versioned (they can update it later and the old version is preserved).
If the conversation discussed choosing PostgreSQL over MySQL:
---
type: decision
status: approved
topic: database-selection
date: 2026-02-19
tags: [database, infrastructure]
---
# Database Selection: PostgreSQL
## Decision
PostgreSQL over MySQL for the new service.
## Reasoning
- Better JSON support (JSONB) for our flexible schema needs
- Superior full-text search (eliminates need for Elasticsearch)
- Row-level security for multi-tenant isolation
## Alternatives Considered
- MySQL: Familiar to team but weaker JSON/FTS support
- MongoDB: Overkill for our relational data patterns
Save to: decisions/database-selection.md
workspace_write (not regular Write tool) — this goes to the persistent workspace, not the local filesystemworkspace_write creates a new version (nothing is overwritten)