CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

wt-switch-create

Create a new worktrunk worktree (optionally in another repo) and switch this session's working directory into it. Use when launching a session that should work in its own worktree.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

The canonical home for this skill is wt-switch-create in max-sixty/worktrunk

SKILL.md
Quality
Evals
Security

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An exceptionally actionable, tightly written operational skill: exact commands, literal error strings, and explicit recovery loops for every failure branch. The only real gaps are minor prose tightening and a maintainer reference (rationale.md) that does not exist in the bundle as provided.

DimensionReasoningScore

Conciseness

The body is dense and operational with no padding and no explanations of concepts Claude already knows; harness-specific facts (stash shared across worktrees, EnterWorktree re-rooting limits) genuinely need stating. Not 5 because a few passages restate the same constraint multiple ways — e.g., the agent-pinned-to-worktree limitation appears twice in step 3, and the repo/branch disambiguation paragraph is a long run-on that could be tightened.

4 / 5

Actionability

Fully executable throughout: exact tool calls (`EnterWorktree({name: "<branch>"})`, `ExitWorktree({action: "keep"})`), a copy-paste CLI command (`wt -C <repo> switch --create <branch> --no-cd --format=json`), the JSON `path` field to read, literal error strings to match ("Already in a worktree session", "Branch <branch> already exists", "Shell cwd was reset"), and four concrete invocation examples.

5 / 5

Workflow Clarity

Steps 1-3 are clearly sequenced with creation-first mandated on every invocation, and each failure path has an explicit feedback loop: stash push/pop with restore-on-failure, branch-exists rerun without `--create`, the `cd`-test recovery split into reachable (pin with `git -C`) vs. Unreachable (ask user to add the directory, then continue), and Denied (report path and ask). Stop-and-ask conditions are stated for unresolvable inputs.

5 / 5

Progressive Disclosure

Well-sectioned single file (argument grammar, examples, What to do, Cleanup, Scope) with all operational content appropriately inline, and maintainer rationale split out to a one-level reference. Not 5 because the referenced `rationale.md (same directory)` is absent from the bundle (no references/ or sibling file exists), leaving a dangling reference; not 3 because structure is clear and nothing that belongs in a separate file is inlined.

4 / 5

Total

18

/

20

Passed

Description

78%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A specific, third-person description with an explicit "Use when" trigger and a clearly named tool niche. Its main limitation is trigger breadth: the single launch-scenario when-clause and sparse synonyms understate the skill's mid-session use cases shown in the body.

Suggestions

Broaden the when-clause to cover mid-session triggers, e.g. "Use when launching a session that should work in its own worktree, or when the user asks to move the current session into a new one mid-task."

Add natural synonyms users would say, such as "git worktree", "create a worktree", or "isolate this work", to the trigger phrasing.

Consider mentioning the uncommitted-work carry-over in the what-clause so the description reflects the skill's full capability set.

DimensionReasoningScore

Specificity

Names several concrete actions — "Create a new worktrunk worktree (optionally in another repo) and switch this session's working directory into it" — three distinct capabilities in third person. Not 5 because coverage is not comprehensive (e.g., stash carry-over and cleanup are body-only); not 3 because it lists more than 1-2 actions.

4 / 5

Completeness

The what is explicit ("Create a new worktrunk worktree ... and switch this session's working directory into it") and the when is an explicit "Use when launching a session that should work in its own worktree" clause. Not 5 because the when-clause covers only the session-launch scenario with a single trigger phrase, while the body also supports mid-session worktree moves; not 3 because both what and when are clearly stated.

4 / 5

Trigger Term Quality

"worktree", "worktrunk", "session", "own worktree", "switch" are natural terms a user needing this skill would say. Not 5 because common variations like "git worktree", "isolate", or mid-session phrasings ("move to a worktree") are missing; not 3 because the dominant natural keyword plus context terms are present and prominent.

4 / 5

Distinctiveness Conflict Risk

"worktrunk worktree", session working-directory switching, and the launch-scenario trigger form a clear niche tied to a named tool with minimal overlap risk. It would not plausibly fire for unrelated skills; the create-plus-switch framing also distinguishes it from a plain switch command.

5 / 5

Total

17

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

max-sixty/worktrunk

Path

plugins/worktrunk/skills/wt-switch-create/SKILL.md

Reviewed

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.