COMPLETE PLUGIN
View an example
Example installation for scribe
Example session using the verified commands above.
Scribe work evidence aggregator — invoicing, project management, timeline tools

Fastest path: hand this line to the coding agent doing the setup. It reads this page and wires the plugin into your harness.
Click anywhere on the prompt to copy.
Choose a runtime, then follow its setup steps.
COMPLETE PLUGIN
Example session using the verified commands above.

This skill should be used when the user asks "what can scribe do", "scribe help", "how does scribe work", "getting started with scribe", "scribe overview", "scribe commands", "what is scribe", "CLI reference", "list scribe commands", "what was my last invoice", "how do I check invoice history", "how do I search my timeline", or wants a general overview of the Scribe system and CLI.

This skill should be used when creating, generating, sending, or managing invoices using the Scribe CLI. Triggers include "create an invoice", "generate an invoice", "invoice a client", "send an invoice", "bill someone", "monthly invoice", "invoice for [month]", "I need an invoice for [amount]", or any mention of invoicing, billing, PDF invoices, client billing, or consulting invoices. Also covers managing clients, checking work summaries for billing periods, and configuring business/payment info.

This skill should be used when the user asks Scribe to turn their work evidence into copy-pasteable prose — "draft a tweet about what I shipped", "write a LinkedIn post from this week", "investor update since November", "turn my week into a thread", "weekly status update for my manager", "changelog blurb from the last release", "narrate what I did on <project>", or any request for social/investor/status copy grounded in the timeline. Covers the three modes (short/long/parts), evidence pulling, noise filtering, and the voice rules.

This skill should be used when setting up Scribe on a fresh device or first-time CLI install — initializing the local database, adding scan roots, discovering and assigning projects, running the first ingest, and configuring business info and clients. Triggers include "set up scribe", "install scribe", "scribe isn't initialized", "first time scribe", "onboard scribe", "scribe init", "fresh install", "new device", "new machine setup", or any request to initialize Scribe locally before invoicing or hub sync.

This skill should be used when managing Scribe projects from the CLI — creating, listing, renaming, deleting projects, linking projects to clients for invoicing, discovering project roots from the filesystem, assigning unassigned roots to project groups, linking git remotes to roots for automatic team project mapping, or viewing project details and statistics. Triggers include "create a project", "show my projects", "link project to client", "discover projects", "assign root", "set remote", "link git remote", "which project is this", "unassigned roots", "project management", or any mention of Scribe project groups, project roots, git remote mapping, or project-client associations.

This skill should be used when the user asks for an activity report or work summary via Scribe — "generate a report", "report on my work", "what did the team get done", "billable vs internal time", "summary of last week/month", "exhaustive report", "PDF report for the client review", or any use of the generateReport capability (CLI `scribe report`, hub MCP, desktop MCP, hub chat, or the /reports page). Covers the three clarifying questions, the defaults, PDF/HTML export, and checking data freshness first.

This skill should be used when an agent wants to read or write Scribe data through MCP tools — "connect Scribe MCP", "scribe MCP tools", "use the Scribe MCP", "search my timeline", "recent activity", "list projects/clients/invoices", "log time to Scribe", "generate a report", "check data freshness", or deciding between the Scribe MCP and the Scribe CLI. Covers the local desktop MCP (localhost, no auth) vs the hosted hub WebMCP (OAuth), and the private-beta admission gate.