CtrlK
BlogDocsLog inGet started
Tessl Logo

prose

OpenProse VM skill pack. Activate on any `prose` command, .prose files, or OpenProse mentions; orchestrates multi-agent workflows.

Install with Tessl CLI

npx tessl i github:attilaczudor/Test --skill prose
What are skills?

68

Does it follow best practices?

Validation for skill structure

SKILL.md
Review
Evals

Evaluation results

90%

20%

Competitive Intelligence Pipeline

Authoring patterns and model tiering

Criteria
Without context
With context

Parallel research

100%

100%

Bounded refinement loop

60%

100%

Orchestrator uses sonnet

100%

100%

Deep analysis uses opus

0%

100%

Haiku not misused

100%

100%

Explicit context declarations

40%

100%

Reusable block defined

0%

0%

Focused agent prompts

100%

100%

Error handling for external calls

100%

100%

No god session

100%

100%

Without context: $0.4709 · 2m 2s · 29 turns · 90 in / 6,887 out tokens

With context: $1.0108 · 2m · 24 turns · 10,152 in / 7,506 out tokens

86%

52%

OpenProse Workspace Upgrade

Workspace migration

Criteria
Without context
With context

.prose/.env created

0%

100%

Key=value format

0%

100%

All keys preserved

0%

100%

state.json noted for removal

100%

100%

Runs directory created

100%

100%

.prose/agents/ created

100%

100%

Migration success messages

0%

0%

Completion status

25%

50%

No data loss

0%

100%

Without context: $0.2354 · 1m 10s · 18 turns · 18 in / 4,355 out tokens

With context: $0.5873 · 1m 28s · 24 turns · 28 in / 5,082 out tokens

100%

72%

OpenProse Team Collaboration Setup

PostgreSQL state configuration and remote resolution

Criteria
Without context
With context

.prose/.env config location

0%

100%

OPENPROSE_POSTGRES_URL key name

0%

100%

Connectivity verification step

33%

100%

Check-before-use ordering

80%

100%

Credentials-in-logs warning

30%

100%

Limited-privilege recommendation

100%

100%

Fallback to filesystem state

37%

100%

Correct --state=postgres flag

0%

100%

Registry short-handle resolution

16%

100%

Docker quick-setup example

0%

100%

Without context: $0.5156 · 2m 6s · 27 turns · 1,797 in / 6,786 out tokens

With context: $0.7152 · 2m 3s · 26 turns · 106 in / 6,612 out tokens

Evaluated
Agent
Claude Code

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.