How do I install the Charted Continue skill?

Run npx skillmds@latest add marmicode/charted-continue in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Charted Continue safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Charted Continue?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Charted Continue free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Charted Continue?

marmicode (@marmicode) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Charted Continue by marmicode — Agent Skill

Name: Charted Continue
Author: marmicode

Charted Continue

Continues Charted Coding work for a PR by routing to scaffold, red, or green based on design doc progress. Use when resuming work on a PR, continuing TDD after a break, or when the user invokes charted continue with a design doc and PR number.

marmicode 0ab3712 3.2 KB Updated Sep 17, 2026

File contents

Context

designDocPath: $ARGUMENTS[0]
prNumber: $ARGUMENTS[1]

Goal

Read ${designDocPath}, inspect the state of PR #${prNumber}, and invoke exactly one downstream skill:

Condition	Skill to invoke
Scaffolding not done	`charted-scaffold`
Next 🚧 test not implemented	`charted-red`
Next 🚧 test already implemented	`charted-green`

Steps

1. Locate the PR section

In ${designDocPath}, find the section for PR #${prNumber} — a heading like # 🚧 PR#N - {title} or # PR#N - {title}.

Read its Tasks and Testing Strategy subsections. The design doc is the single source of truth for what belongs to this PR.

2. Check scaffolding

Scaffolding is done when all of the following hold for PR #${prNumber}:

Test file(s) exist for the PR's Testing Strategy scenarios.
Each ### 🚧 {test_title} scenario has a matching it.todo("{test_title}", ...) in a test file (test names are lower case).
WIP production files referenced in Tasks exist (classes/components tagged @deprecated 🚧 work in progress or equivalent stubs).

Scaffolding is not done if any expected test file or WIP source file is missing, or if it.todo stubs with design-doc step comments do not yet exist.

If scaffolding is not done → read and follow the charted-scaffold skill with ${designDocPath} and ${prNumber}. Stop.

3. Find the next pending test

In the PR's Testing Strategy, pick the first heading still marked ### 🚧 {test_title} (not ### ✅).

If every test is ### ✅, report that PR #${prNumber} is complete and stop.

4. Locate the matching test file

Search the project for the it.todo whose name matches the pending test title (case-insensitive). Prefer the file under the component or unit named in the nearest ## {Component} heading above the scenario.

Record the path as testFilePath.

5. Classify the test

Open testFilePath and inspect the it.todo body for the pending test:

Not implemented — body is empty or contains only step-by-step comments copied from the design doc (no expect, inject, mount helpers, or other executable test code).
Implemented — body contains actual test code beyond comments.

6. Invoke the downstream skill

Not implemented → read and follow the charted-red skill with ${designDocPath} and ${testFilePath}.

Implemented → read and follow the charted-green skill with ${designDocPath} and ${testFilePath}.

Rules

Invoke one downstream skill per run — never scaffold, red, and green in the same turn.
Do not skip ahead to a later 🚧 test while an earlier one is still pending.
Do not implement production code or enable tests directly — delegate to the downstream skill.
When reading downstream skills, follow their instructions fully, including Wallaby usage and design-doc checkbox updates.

---name: charted-continuedescription: Continues Charted Coding work for a PR by routing to scaffold, red, or green based on design doc progress. Use when resuming work on a PR, continuing TDD after a break, or when the user invokes charted continue with a design doc and PR number.---# Context- designDocPath: $ARGUMENTS[0]- prNumber: $ARGUMENTS[1]# GoalRead `${designDocPath}`, inspect the state of PR #${prNumber}, and invoke exactly one downstream skill:| Condition | Skill to invoke || -------------------------------- | ------------------ || Scaffolding not done | `charted-scaffold` || Next 🚧 test not implemented | `charted-red` || Next 🚧 test already implemented | `charted-green` |# Steps## 1. Locate the PR sectionIn `${designDocPath}`, find the section for PR #${prNumber} — a heading like `# 🚧 PR#N - {title}` or `# PR#N - {title}`.Read its **Tasks** and **Testing Strategy** subsections. The design doc is the single source of truth for what belongs to this PR.## 2. Check scaffoldingScaffolding is **done** when all of the following hold for PR #${prNumber}:- Test file(s) exist for the PR's Testing Strategy scenarios.- Each `### 🚧 {test_title}` scenario has a matching `it.todo("{test_title}", ...)` in a test file (test names are lower case).- WIP production files referenced in Tasks exist (classes/components tagged `@deprecated 🚧 work in progress` or equivalent stubs).Scaffolding is **not done** if any expected test file or WIP source file is missing, or if `it.todo` stubs with design-doc step comments do not yet exist.**If scaffolding is not done** → read and follow the `charted-scaffold` skill with `${designDocPath}` and `${prNumber}`. **Stop.**## 3. Find the next pending testIn the PR's **Testing Strategy**, pick the **first** heading still marked `### 🚧 {test_title}` (not `### ✅`).If every test is `### ✅`, report that PR #${prNumber} is complete and **stop**.## 4. Locate the matching test fileSearch the project for the `it.todo` whose name matches the pending test title (case-insensitive). Prefer the file under the component or unit named in the nearest `## {Component}` heading above the scenario.Record the path as `testFilePath`.## 5. Classify the testOpen `testFilePath` and inspect the `it.todo` body for the pending test:- **Not implemented** — body is empty or contains only step-by-step comments copied from the design doc (no `expect`, `inject`, mount helpers, or other executable test code).- **Implemented** — body contains actual test code beyond comments.## 6. Invoke the downstream skill**Not implemented** → read and follow the `charted-red` skill with `${designDocPath}` and `${testFilePath}`.**Implemented** → read and follow the `charted-green` skill with `${designDocPath}` and `${testFilePath}`.# Rules- Invoke **one** downstream skill per run — never scaffold, red, and green in the same turn.- Do not skip ahead to a later 🚧 test while an earlier one is still pending.- Do not implement production code or enable tests directly — delegate to the downstream skill.- When reading downstream skills, follow their instructions fully, including Wallaby usage and design-doc checkbox updates.

Frequently asked questions