/zoxide
Resolve a named repo, project, or directory to its most likely local path using zoxide's frecency database, so you can jump in and work. Use when the user references a repo/project by name without a path ("work on the honeycomb cli", "open my dotfiles", "the aws provider repo"),
$ npx -y skills add bendrucker/claude --skill zoxide --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/zoxide
Context preview
The summary Claude sees to decide when to auto-load this skill.
Resolve a named repo, project, or directory to its most likely local path using zoxide's frecency database, so you can jump in and work. Use when the user references a repo/project by name without a path ("work on the honeycomb cli", "open my dotfiles", "the aws provider repo"),
SKILL.md
zoxide.SKILL.mdname: zoxide
description: Resolve a named repo, project, or directory to its most likely local path using zoxide's frecency database, so you can jump in and work. Use when the user references a repo/project by name without a path ("work on the honeycomb cli", "open my dotfiles", "the aws provider repo"), or when you need to locate a directory before operating in it.
argument-hint: "[keywords]"
allowed-tools:
- BashZoxide Repo Resolver
Turn a human reference into an absolute local path. When the user names a repo, project, or directory by name without giving a path, zoxide's frecency database says which directory *this user* visits most under that name. That is the answer.
Invoke whenever:
- The user names a repo/project/directory without a path and you need to act in it ("work on the honeycomb cli", "open my dotfiles", "the aws provider repo").
- A path the user gave doesn't exist and you must infer what they meant.
Resolve
Raw `zoxide query` opens the database read-write to age scores and prune missing dirs. Under the sandbox it prints the correct path but also writes `could not write to database` to stderr and **sometimes returns rc=1 on a correct answer**. So copy `db.zo` into `$TMPDIR` once and query the copy through `_ZO_DATA_DIR`:
ZO_SRC="${_ZO_DATA_DIR:-$HOME/Library/Application Support/zoxide}" # honors a relocated db
[ -d "$ZO_SRC" ] || ZO_SRC="${XDG_DATA_HOME:-$HOME/.local/share}/zoxide"
ZO_TMP="${TMPDIR:-/tmp}/zoxide.$$"; mkdir -p "$ZO_TMP"; cp "$ZO_SRC/db.zo" "$ZO_TMP/db.zo"
z() { _ZO_DATA_DIR="$ZO_TMP" zoxide query "$@"; }
z honeycomb # best match: /Users/.../honeycomb-cli
z bendrucker claude # multi-keyword narrows, in path order
z -l -s provider # ranked candidates + scores, for disambiguationReuse the same `$ZO_TMP` copy across queries within one activation. `no match found` (rc=1 from the copy) means zoxide has never tracked such a directory. Say so rather than guessing a path.
Disambiguate
- One clear best match → use it.
- If the top match is a broad parent (e.g. `/Users/ben/src`) rather than a specific repo, add keywords or run `z -l -s <kw>` and pick the most specific, highest-scored repo.
- If the top two candidates are close in score and are genuinely different targets, present both and let the user pick rather than silently choosing.
After Resolving
Hand the absolute path to the caller and operate there, with absolute paths or the `worktrunk:wt-switch-create` skill for a worktree. This skill only reads zoxide. It does not run `zoxide add`/`remove`/`edit`/`import` and does not mutate the ranking. The default query already filters to directories that still exist, so the resolved dir is one zoxide believes is present. If a later command shows it's gone, re-resolve with more keywords.
Scope
zoxide ranks directories by frecency and exposes no per-visit timestamps, so this skill answers "which directory," not "when." For shell-command history ("what/when did I run"), use `atuin:history`.
Read more
name: zoxide
description: Resolve a named repo, project, or directory to its most likely local path using zoxide's frecency database, so you can jump in and work. Use when the user references a repo/project by name without a path ("work on the honeycomb cli", "open my dotfiles", "the aws provider repo"), or when you need to locate a directory before operating in it.
argument-hint: "[keywords]"
allowed-tools:
- BashZoxide Repo Resolver
Turn a human reference into an absolute local path. When the user names a repo, project, or directory by name without giving a path, zoxide's frecency database says which directory *this user* visits most under that name. That is the answer.
Invoke whenever:
- The user names a repo/project/directory without a path and you need to act in it ("work on the honeycomb cli", "open my dotfiles", "the aws provider repo").
- A path the user gave doesn't exist and you must infer what they meant.
Resolve
Raw `zoxide query` opens the database read-write to age scores and prune missing dirs. Under the sandbox it prints the correct path but also writes `could not write to database` to stderr and **sometimes returns rc=1 on a correct answer**. So copy `db.zo` into `$TMPDIR` once and query the copy through `_ZO_DATA_DIR`:
ZO_SRC="${_ZO_DATA_DIR:-$HOME/Library/Application Support/zoxide}" # honors a relocated db
[ -d "$ZO_SRC" ] || ZO_SRC="${XDG_DATA_HOME:-$HOME/.local/share}/zoxide"
ZO_TMP="${TMPDIR:-/tmp}/zoxide.$$"; mkdir -p "$ZO_TMP"; cp "$ZO_SRC/db.zo" "$ZO_TMP/db.zo"
z() { _ZO_DATA_DIR="$ZO_TMP" zoxide query "$@"; }
z honeycomb # best match: /Users/.../honeycomb-cli
z bendrucker claude # multi-keyword narrows, in path order
z -l -s provider # ranked candidates + scores, for disambiguationReuse the same `$ZO_TMP` copy across queries within one activation. `no match found` (rc=1 from the copy) means zoxide has never tracked such a directory. Say so rather than guessing a path.
Disambiguate
- One clear best match → use it.
- If the top match is a broad parent (e.g. `/Users/ben/src`) rather than a specific repo, add keywords or run `z -l -s <kw>` and pick the most specific, highest-scored repo.
- If the top two candidates are close in score and are genuinely different targets, present both and let the user pick rather than silently choosing.
After Resolving
Hand the absolute path to the caller and operate there, with absolute paths or the `worktrunk:wt-switch-create` skill for a worktree. This skill only reads zoxide. It does not run `zoxide add`/`remove`/`edit`/`import` and does not mutate the ranking. The default query already filters to directories that still exist, so the resolved dir is one zoxide believes is present. If a later command shows it's gone, re-resolve with more keywords.
Scope
zoxide ranks directories by frecency and exposes no per-visit timestamps, so this skill answers "which directory," not "when." For shell-command history ("what/when did I run"), use `atuin:history`.
My personal plugin marketplace for Claude Code, Anthropic's AI coding assistant.
Repo: bendrucker/claude
Other skills on bendrucker-claude.
- /agent-ideas
Harvest agent-tooling ideas from prominent developers.
Open skill - /cleye
Type-safe CLI argument parsing with cleye, the standard parser for this repo's Bun scripts. Use when writing or editing any script that takes arguments (flags, positional parameters, subcommands, --help) instead of reading existing scripts for the pattern.
Open skill - /coverage
Measure Bun test coverage and close gaps on a specific file. Use when adding or editing tests, when asked about coverage, or when the PostToolUse coverage hook reports uncovered lines.
Open skill - /activity
Report real device usage from ActivityWatch. Covers per-app time, window titles, and active vs idle spans. Use when asked "what apps did I use", "how long was I in X", "what did I work on today", "how much was I active vs idle", or to mine usage patterns for automation.
Open skill - /history
Report shell history from atuin's local capture. Covers what commands ran, when, where, and how they exited. Use when asked "what commands did I run", "what was I working on in the terminal", "have I ever run X", "how do I usually invoke X", or about recent shell activity,
Open skill - /bun
Bun runtime patterns. Use when running bun commands, working with package.json/bun.lock, writing TypeScript scripts under Bun, or developing Claude Code plugins.
Open skill

