plan-converter_skill

This skill converts any planning artifact into .task/.json multi-file format for unified execution and traceable task management.
  • TypeScript

1.3k

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill catlog22/claude-code-workflow --skill plan-converter

  • SKILL.md14.9 KB

Overview

This skill converts planning, analysis, and brainstorm artifacts into the standardized .task/*.json multi-file format consumed by unified-execute-with-file. It auto-detects common inputs (roadmap.jsonl, plan-note.md, conclusions.json, synthesis.json, legacy JSONL) and emits one validated TASK-{id}.json file per task in a .task/ directory or specified output path. The converter focuses on producing complete convergence fields and dependency metadata so tasks are execution-ready.

How this skill works

The converter parses CLI arguments to locate the input and optional -o output path, detects the input format, and extracts raw records using format-specific parsers. It transforms each source record into a unified task schema (identity, dependencies, convergence, files, implementation hints, context) and runs convergence-quality validation with auto-fix suggestions before writing cleaned JSON files to .task/*.json. A final summary reports record counts and any unresolved quality issues.

When to use it

  • You have a roadmap.jsonl, plan-note.md, conclusions.json or synthesis.json and need execution-ready task files.
  • Standardizing disparate planning artifacts into a single per-task JSON format for automated orchestration.
  • Preparing tasks with explicit convergence criteria and verification steps for multi-agent workflows or CI-driven execution.
  • Migrating legacy JSONL or ad-hoc markdown plans into a unified schema consumed by CLI tooling.
  • Generating per-task files to populate a .task/ directory for downstream runners or review.

Best practices

  • Provide session metadata (YAML frontmatter) in plan-note.md to preserve source.session_id and provenance.
  • Ensure task descriptions include scope and file change hints so convergence can be generated accurately.
  • Review and refine auto-generated convergence.criteria and verification entries to avoid vague language.
  • Supply roadmap.jsonl with unified fields (title, description, convergence) when possible to minimize post-conversion edits.
  • Use the -o flag to control output location and avoid accidental overwrites of existing .task/ folders.

Example use cases

  • Convert a req-plan roadmap.jsonl into .task/ for automated multi-agent execution and tracking.
  • Turn brainstorm synthesis.json into prioritized IDEA-* task files with feasibility-based effort estimates.
  • Extract recommendations from conclusions.json into discrete tasks with generated definition_of_done and verification.
  • Parse collaborative plan-note.md into per-domain TASK-NNN.json files, including file change directives and conflict risk.
  • Normalize legacy tasks.jsonl or generic JSONL into the unified task schema for downstream CI or orchestrators.

FAQ

roadmap.jsonl, tasks.jsonl (legacy), plan-note.md, conclusions.json, synthesis.json, generic .jsonl and .json with detectable shapes.

Where are outputs written?

By default outputs go to a .task/ subfolder next to the input file; use -o <path> to set a custom output directory.

How does the tool handle missing convergence details?

It auto-generates convergence.criteria, verification, and definition_of_done from title, description, and file-change context, and flags items that need manual refinement.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
plan-converter skill by catlog22/claude-code-workflow | VeilStrat