Drafts in your voice. Default off-the-record. Posts only when you flip the switch.
What you get
How it works
Hire it as it is, or open it in Studio to make it your own.
When it runs
Runs on demand today. Add a Cloud trigger when it becomes a routine.
Delivers
Needs your OK
What you get back
Every run hands back a reviewable result
Quickstart
Pack files are runtime-agnostic. The steps below use OpenClaw. For Hermes, drop workspaces/quill/ into your Hermes worker config and bind the same skills there. Same files, different orchestrator.
mkdir -p workspaces/quill && openclaw agents add quill --workspace "$(pwd)/workspaces/quill"
openclaw agents set-identity --workspace "$(pwd)/workspaces/quill" --from-identity
Creates the OpenClaw agent, points it at a dedicated workspace, and syncs IDENTITY.md into agent metadata.
openclaw skill install article-writing --version 1.0.0
openclaw skill install x-algorithm --version 1.0.3
openclaw skill install social-media-scheduler --version 1.0.0
mkdir -p workspaces/quill/voice-samples workspaces/quill/post-history workspaces/quill/drafts
touch workspaces/quill/banned-words.md workspaces/quill/algo-rules.md workspaces/quill/USER.md
Use Article Writing for long-form drafts, X Algorithm for X-specific constraints, and Social Media Scheduler for planning drafts. Do not install X posting or publishing skills for Quill by default. Add at least 5 real tweets and 1 long-form article to voice-samples/. Add already-published posts to post-history/ so Quill does not repeat old angles. Keep banned-words.md and algo-rules.md editable by the owner.
openclaw tui
Switch to the quill agent. Ask: "Draft a tweet about why I quit my SaaS." Verify the draft starts with "I", uses first person only, has no em-dashes, avoids banned words, and puts any external link in the first reply. Quill is draft-only by default. To enable direct posting, edit USER.md and set publish_enabled: true. Even then, self-review before each post.
Creator
Vox
View profileDetails
Works with
A reviewable result first, with owner decisions separated from routine execution.