OpenProse
OpenProse is a portable, markdown-first workflow format for orchestrating AI sessions. In Fased, it ships as a plugin that installs an OpenProse skill pack and a/prose slash command. Programs live in .prose files and can spawn multiple
agent sessions with explicit control flow.
Official site: https://www.prose.md
What it can do
- Multi-agent research + synthesis with explicit parallelism.
- Repeatable approval-safe workflows (code review, incident triage, content pipelines).
- Reusable
.proseprograms you can run across supported agent runtimes.
Install + enable
Bundled plugins are disabled by default. Enable OpenProse:fased plugins install ./extensions/open-prose
Related docs: Plugins, Plugin manifest, Skills.
Slash command
OpenProse registers/prose as a user-invocable skill command. It routes to
the OpenProse VM instructions and uses the normal Fased tool/runtime surface
under the hood.
Common commands:
Example: a simple .prose file
File locations
OpenProse keeps state under.prose/ in your workspace:
State modes
OpenProse supports multiple state backends:- filesystem (default):
.prose/runs/... - in-context: transient, for small programs
- sqlite (experimental): requires
sqlite3binary - postgres (experimental): requires
psqland a connection string
- sqlite/postgres are opt-in and experimental.
- postgres credentials flow into agent-session logs; use a dedicated, least-privileged DB.
Remote programs
/prose run <handle/slug> resolves to https://p.prose.md/<handle>/<slug>.
Direct URLs are fetched as-is. This uses the web_fetch tool (or exec for POST).
Fased runtime mapping
OpenProse programs map to Fased primitives:
If your tool allowlist blocks these tools, OpenProse programs will fail. See Skills config.
Security + approvals
Treat.prose files like code. Review before running. Use Fased tool
allowlists and approval gates to control side effects.
For scheduled or recurring work, use Tasks. For one
structured model step inside a workflow, use LLM Task.