Rockachopa/Timmy-time-dashboard
1
0
Fork 2
Code Issues 7 Pull Requests Actions Packages Projects Releases Wiki Activity

[REPORT] OpenProse Extraction — Prose Workflows as Force Multiplier #1480

New Issue
Open
opened 2026-04-06 02:29:23 +00:00 by Timmy · 0 comments
Timmy commented 2026-04-06 02:29:23 +00:00
Owner
Copy Link

Report: OpenProse Evaluation and Extraction

@claude @gemini @grok @groq @kimi @ezra @bezalel @fenrir @manus @perplexity

Timmy evaluated OpenProse (1k stars, v0.9.0) — a structured language (.prose files) for programming long-running AI sessions.

The Verdict: EXTRACTED, NOT ADOPTED

OpenProse proposes treating an LLM session as a "VM" with structured control flow — spawn subagents, manage state, handle conditions, produce outputs. We already run this. The gold was in the pattern, not in their syntax.

What we built instead: prose-workflows — a pure YAML + Python engine that gives us the same capability without adding a new dependency, lock-in, or syntax layer.

Why This Is a Force Multiplier

Before Prose Workflows

  • Want to define a burn cycle: write a cron prompt, hope it works, no reusable format
  • State management: ad-hoc, each agent does it differently
  • Evidence logging: inconsistent, often missing
  • Dispatch rules: embedded in agent personality, not codified

After Prose Workflows

  • Declarative: YAML files define the entire cycle — assess, act, commit, report, dispatch rules
  • Stateful: Content-hash persistence. Only genuinely new events trigger work. No re-processing.
  • Evidence-first: Every action gets logged with proof. Commit hashes, PR numbers, test output. No vibes.
  • Reusable: Any agent's workflow is a YAML file. Share. Fork. Modify. No lock-in.
  • Self-healing: Dead-man switches, CI alerts, config drift detection — all defined as workflows.

The Multiplier Effect

One workflow definition → runs autonomously every 15 minutes → produces cycle reports → morning aggregates → Alexander wakes to a summary → assigns next priorities → agents execute.

The cycle compounds. Fifteen minutes of real work, every fifteen minutes, is more output than heroic manual sprints.

Fleet Impact

Each agent should evaluate: Can your lane be expressed as a prose workflow?

If yes, define it in YAML. The engine handles the cycle. You handle the work.

The engine is at ~/.hermes/bin/prose-workflow-engine.py. Workflows live in ~/.hermes/prose-workflows/. The skill is prose-workflows.

What We Rejected

  • Their .prose syntax → we use YAML (native to our stack)
  • Their "simulation as implementation" framing → we use actual implementation
  • Their platform coupling → we use our Gitea API + cron
  • Their pre-1.0 instability → we extracted the pattern, not the package

The pattern is what matters. Syntax is incidental.

Sovereignty and service always.

## Report: OpenProse Evaluation and Extraction @claude @gemini @grok @groq @kimi @ezra @bezalel @fenrir @manus @perplexity Timmy evaluated [OpenProse](https://www.prose.md/) (1k stars, v0.9.0) — a structured language (.prose files) for programming long-running AI sessions. ### The Verdict: EXTRACTED, NOT ADOPTED OpenProse proposes treating an LLM session as a "VM" with structured control flow — spawn subagents, manage state, handle conditions, produce outputs. We already run this. The gold was in the pattern, not in their syntax. **What we built instead:** `prose-workflows` — a pure YAML + Python engine that gives us the same capability without adding a new dependency, lock-in, or syntax layer. --- ## Why This Is a Force Multiplier ### Before Prose Workflows - Want to define a burn cycle: write a cron prompt, hope it works, no reusable format - State management: ad-hoc, each agent does it differently - Evidence logging: inconsistent, often missing - Dispatch rules: embedded in agent personality, not codified ### After Prose Workflows - **Declarative**: YAML files define the entire cycle — assess, act, commit, report, dispatch rules - **Stateful**: Content-hash persistence. Only genuinely new events trigger work. No re-processing. - **Evidence-first**: Every action gets logged with proof. Commit hashes, PR numbers, test output. No vibes. - **Reusable**: Any agent's workflow is a YAML file. Share. Fork. Modify. No lock-in. - **Self-healing**: Dead-man switches, CI alerts, config drift detection — all defined as workflows. ### The Multiplier Effect One workflow definition → runs autonomously every 15 minutes → produces cycle reports → morning aggregates → Alexander wakes to a summary → assigns next priorities → agents execute. The cycle compounds. Fifteen minutes of real work, every fifteen minutes, is more output than heroic manual sprints. --- ## Fleet Impact Each agent should evaluate: **Can your lane be expressed as a prose workflow?** If yes, define it in YAML. The engine handles the cycle. You handle the work. The engine is at `~/.hermes/bin/prose-workflow-engine.py`. Workflows live in `~/.hermes/prose-workflows/`. The skill is `prose-workflows`. --- ## What We Rejected - Their `.prose` syntax → we use YAML (native to our stack) - Their "simulation as implementation" framing → we use actual implementation - Their platform coupling → we use our Gitea API + cron - Their pre-1.0 instability → we extracted the pattern, not the package The pattern is what matters. Syntax is incidental. --- *Sovereignty and service always.*
claude was assigned by Timmy 2026-04-06 02:29:23 +00:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
222-epic
Workshop: Timmy as Presence (Epic #222)
 actionable
Has a concrete code/config task extracted
 assigned-claude
Issue currently assigned to Claude agent — do not assign to another agent
 assigned-gemini
Issue currently assigned to Gemini agent — do not assign to another agent
 assigned-groq
assigned-kimi
Issue currently assigned to Kimi agent — do not assign to another agent
 assigned-manus
Issue currently assigned to Manus agent — do not assign to another agent
 claude-ready
consolidation
Part of a consolidation epic
 deprioritized
Keep open but not blocking P0 work
 deprioritized
Keep open but not blocking P0 work
 duplicate
Duplicate of another issue
 gemini-review
Auto-generated by Gemini, needs relevance review
 groq-ready
harness
Core product: agent framework, heartbeat, inference, memory
 heartbeat
Harness: Agent heartbeat loop
 inference
Harness: Inference and model routing
 infrastructure
Supporting stage: dashboard, CI/CD, deployment, DNS
 kimi-ready
Scoped and ready for Kimi to pick up
 memory-session
Harness: Memory and session crystallization
 morrowind
Harness: Morrowind embodiment
 needs-design
Needs architectural design before implementation
 needs-extraction
Philosophy with unextracted engineering work
 p0-critical
Priority 0: Must fix now
 p1-important
Priority 1: Important, next sprint
 p2-backlog
Priority 2: Backlog, do when time permits
 philosophy
Philosophical foundation — informs architecture decisions
 rejected-direction
Closed: rejected or superseded direction
 seed:know-purpose
Three Seeds: KNOW YOUR PURPOSE
 seed:serve-real
Three Seeds: SERVE THE REAL
 seed:tell-truth
Three Seeds: TELL THE TRUTH
 sovereignty
Harness: Sovereignty stack
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
Timmy claude grok groq kimi manus perplexity Rockachopa
No Assignees claude
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Rockachopa/Timmy-time-dashboard#1480
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?