j.playbook-new

Guides authoring of project-local playbook JSON files with catalog-validated inputs and post-write verification.

Updated Aug 3, 2026
One-click install
npx skills add https://github.com/samwelmunga/jenga-npm --skill j-playbook-new-samwelmunga
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: j.playbook-new
Source: https://github.com/samwelmunga/jenga-npm/tree/main/.agents/skills/j-playbook-new
Command: npx skills add https://github.com/samwelmunga/jenga-npm --skill j-playbook-new-samwelmunga

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Hand-authoring a project-local playbook JSON file is error-prone: ids can collide with existing playbooks, skill names can be misspelled, and required fields like keywords and examples are easy to omit, leaving a file that fails schema validation or never matches natural-language requests. This Skill walks you through each required field, validates every input against the live playbook and skill catalogs, and confirms the written file actually loads before reporting success. ## Core Features & Use Cases - Guided field-by-field wizard: Prompts for id, name, description, keywords, examples, and an ordered list of at least two skills, re-prompting with verbatim reasons whenever validation fails. - Live catalog validation: Checks id slug format and uniqueness against the merged built-in plus project playbook catalog, and validates each skill name against the generated skill catalog used by natural-language matching. - Self-validating writes: Writes project/.playbooks/<id>.json via a deterministic script, then re-validates the file through the playbook loader before declaring success. - Use Case: You want a custom release workflow chaining your plan, implement, and test skills. Run the wizard, answer the prompts, and receive a confirmed-loadable playbook invocable by id or matched from free-text requests. ## Quick Start Ask the agent to help you create a new project-local playbook and follow the guided prompts for id, name, description, keywords, examples, and the ordered skill chain.

Frequently Asked Questions about j.playbook-new

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

FAQPage Schema
How do I create a custom playbook for my project?▼

Run the guided wizard, which prompts for an id, name, description, keywords, examples, and an ordered list of at least two skills. It validates each input against the live catalogs, writes project/.playbooks/<id>.json, and confirms the file loads before reporting success.

How are playbook skill names validated?▼

Each skill name is checked against the generated skill catalog produced by load-nl-catalog.sh, the same source the natural-language matcher uses. Invalid names return a verbatim reason and the wizard re-prompts rather than guessing a correction.

Can a playbook include conditional or composition steps?▼

The wizard authors only plain ordered bare-string skill chains, with no forward_from, resolve, conditional, or composition StepObject fields. To use those fields, hand-edit the written project/.playbooks/<id>.json file afterward; the schema supports them identically.

Why does my playbook id get rejected during authoring?▼

Ids must match the slug pattern ^[a-z0-9]+(-[a-z0-9]+)*$ and must not collide with any loaded playbook or an existing file at project/.playbooks/<id>.json, even if that file is currently invalid. Choose a different id or edit the existing file directly.

What happens if the written playbook fails validation?▼

The wizard runs the playbook loader's lookup on the new id after writing. If the status is invalid or not_found, it reports the verbatim reason and never claims success, surfacing schema defects or environment problems instead of hiding them.