Agent0 is an AI agent framework built on Obsidian and Claude Code. It gives you 7 specialized AI agents powered by 83 reusable skills, with built-in governance, persistent memory, and a vault that catches every idea and ships the ones that matter.

Do I need to know how to code to use Agent0?

No. The Starter tier uses copy-paste prompts with step-by-step instructions. The Pro tier runs one command. The Custom tier is fully set up for you.

How is Agent0 different from other AI agent frameworks?

Agent0 runs on Obsidian and Claude Code with your data staying local. No hosted servers, no monthly infrastructure costs. The reflect/evolve layer (seeing your thinking patterns over time) is something no other framework offers.

When I am here, I trigger. When I am not, it runs itself.

Slash commands do the work during the day. A scheduled heartbeat does the work at night. Nothing compounds by accident.

The human part

Slash commands

Twenty-plus commands cover the loop from brain dump to ship. Type one. The agents pick it up.

Daily flow

/daily

Start the day. Brain dump, capture processing, agent briefing.

/tasks

Review what agents proposed. Approve, reject with feedback, or skip.

/get-shit-done

Execute approved tasks. Only approved ones. Nothing pending.

/tldr

End the session. Summarize, save, update memory, push to git.

Capture and ship

/capture

Extract from TikTok, YouTube, Twitter, GitHub into the vault.

/distribute

Turn an article into platform-native LinkedIn and X drafts.

/publish

Push to Webflow and queue posts in Postiz. Human review required.

/sync

Pull agent PRs, auto-merge the safe ones, push local changes.

Reflection and thinking

/reflect

Weekly and monthly reflection. Life balance, wins, misses.

/trace

Track how one idea evolved across the vault over time.

/connect

Find unexpected bridges between two topics.

/challenge

Pressure-test beliefs. Surface contradictions and weak assumptions.

/emerge

Find clusters of thinking ready to become projects.

/drift

Surface recurring themes I am circling without realizing.

Maintenance

/maintain

Vault cleanup. Fix broken links, update frontmatter, surface orphans.

/standup

Cross-domain briefing across every active project.

/context

Load full life plus work state into the session.

/changelog

Auto-logged. Tracks system-level changes to the framework.

The autonomous part

Scheduled agents

One consolidated heartbeat runs three times a day via the Claude Code RemoteTrigger API. It plays different roles based on the time. Two agents share the work. Radar goes out. Cortex stays in.

ScoutRadar

Every run

Goes out. Monitors flagged GitHub repos. Flags unprocessed captures in daily notes. External API calls allowed.

Task ProposalsCortex

Every run

Stays in. Scans tasks, ideas, daily notes for actionable items. Proposes up to 5 tasks into the Backlog column.

ChronicleCortex

10pm ET

Appends a Chronicle Entry to today’s daily note. Reads the day’s sessions and recent commits. Updates memory.

MirrorCortex

Sunday midnight ET

Weekly reflection. Life balance scorecard, wins, misses. Saves to daily/reflections/.

CompassCortex

1st of month, midnight ET

Monthly reflection. Goal progress, decisions logged, focus for next month. Saves to daily/reflections/.

How one run works

Same heartbeat. Different roles stack based on the time.

01Heartbeat triggers via RemoteTrigger API.
02Agent checks current UTC time to figure out which roles to play.
03Scout checks monitored repos and flags captures that need processing.
04Task Proposals scans the vault. Proposes up to 5 tasks into the Backlog.
05Chronicle runs at 10pm ET. Writes the day into today’s daily note. Updates memory.
06Mirror runs Sunday. Weekly reflection in daily/reflections/.
07Compass runs the 1st. Monthly reflection in daily/reflections/.
08All changes commit to agent/cortex-YYYY-MM-DD-HHMM. PR opened.
09/sync auto-merges the safe ones. Nothing ships without governance.