charting

Plan multi-session efforts by mapping decisions and tasks as tickets on an issue tracker.

2|Updated Jul 24, 2026
One-click install
npx skills add https://github.com/konstantinos-malavazos/claude-code-playbook --skill charting-konstantinos-malavazos
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: charting
Source: https://github.com/konstantinos-malavazos/claude-code-playbook/tree/main/templates/skills/charting
Command: npx skills add https://github.com/konstantinos-malavazos/claude-code-playbook --skill charting-konstantinos-malavazos

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Large efforts often have a clear destination but no visible route, making them impossible to plan in a single session. This Skill turns that fog into a map of tickets on your issue tracker, then resolves them one per session until nothing is left to decide. ## Core Features & Use Cases - Map-based planning: Creates one map issue whose child tickets are either decisions (no files written) or makes (no new decisions), each sized to one agent session. - Tracker-agnostic verbs: Communicates with any issue tracker through intent verbs (create, claim, link, frontier) resolved by a single installed adapter, never raw CLI calls. - Fog management: Keeps unshaped work in a Not yet specified section and graduates it into tickets only when the question becomes precise. - Use Case: You inherit a migration spanning multiple repos with no clear plan. Run charting to grill the open decisions breadth-first, create a map with research, grilling, and make tickets wired by blocking edges, then work the frontier one ticket per session until the map closes. ## Quick Start Ask the AI to chart this effort into a map of decision and make tickets on the issue tracker, starting by settling the destination together.

Frequently Asked Questions about charting

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I plan a project that is too big for one session?

Chart it: settle the destination with the human, grill the open questions breadth-first, then create a map issue with child tickets sized to one session each. Work the frontier one ticket per session until nothing is left to decide.

What is the difference between a decision ticket and a make ticket?

A decision ticket writes no files and ends when the human and agent agree; a make ticket decides nothing new and ends when the file is on disk. They never share a ticket because their endings are unpredictable in different ways.

Does charting work with any issue tracker?

Yes. Charting never names a tracker or writes raw CLI calls; it states intents as verbs like create, claim, and frontier, which a single installed adapter doc at ~/.claude/tracker.md translates to your tracker.

When should I not use charting for a task?

Skip charting when the work can already be planned in one pass. If breadth-first grilling surfaces no fog, the effort fits one session and no map is needed.

Why does every resolution comment need a leading one-line gist?

The gist makes rebuilding the map's Decisions so far section a concatenation of first lines instead of re-summarizing every closed ticket, which would be expensive and inconsistent across runs.