vault/Operations/Claude/skills/system/chief-of-staff/SKILL.md
Use when user asks about their emails, tasks, calendar, project status, memory, or wants the AI Chief of Staff to handle a request
npx skillsauth add eddowding/cortex chief-of-staffInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Quick access to your AI Chief of Staff agent for context retrieval and task execution.
This skill invokes the {SYSTEM_NAME} agent with user arguments. {SYSTEM_NAME} has access to Gmail, TickTick, memory systems, and workspace context.
Invoke immediately when user asks about:
Natural language triggers:
// Invoke agent with Task tool
Task({
subagent_type: "{SYSTEM_NAME}",
prompt: `${userQuery}
Provide ACTUAL DATA/RESULTS, not metadata about what was done.
Ultrathink for deeper reasoning.`,
description: "Query via {SYSTEM_NAME} agent"
})
| User Query | What {SYSTEM_NAME} Does | |------------|----------------| | "Did I email X?" | Search Gmail sent, relevant docs, memory | | "What's hot right now?" | Scan recent git, email, tasks - surface priorities | | "Check email for X" | Gmail search with context | | "Status on project X" | Check project file + recent activity |
{SYSTEM_NAME} has access to:
Always request:
| Mistake | Fix |
|---------|-----|
| Searching yourself instead of invoking {SYSTEM_NAME} | Let {SYSTEM_NAME} handle - has full context |
| Forgetting to ask for actual data | Request results first, not metadata |
| Not using "ultrathink" | Always include for better reasoning |
| Using {"tags": ["tag"]} for TickTick filter | Use {"tag_label": "tag"} - old syntax broken |
tools
Dispatch to the right problem-solving technique based on how you're stuck.
testing
Interactive brainstorming and advisory mode that helps think through questions, critique approaches, propose alternatives, and reach optimal decisions.
tools
Find one insight that eliminates multiple components - 'if this is true, we don't need X, Y, or Z'
testing
Test at extremes (1000x bigger/smaller) to expose fundamental truths hidden at normal scales.