How do I install the Paw skill?

Run npx skillmds@latest add octaviantocan/paw 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.

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 Paw?

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.

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

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

Paw by octaviantocan — Agent Skill

Name: Paw
Author: octaviantocan

File contents

paw

Use paw when you need the supported Pawrrtal CLI surface. This CLI is a Bun-backed Effect v4 package at packages/paw-cli; do not route new work through the deleted Python CLI.

Commands

Command	Aliases	Summary
`paw doctor`	-	Check local Paw CLI readiness
`paw context`	whoami	Print the active CLI context without secrets
`paw completions <shell>`	-	Generate shell completions

Common Usage

just paw --help
just paw doctor --json
just paw context --json
just paw whoami --plain
just paw completions zsh
bun run --filter @pawrrtal/cli start -- doctor

Global Options

Option	Meaning
`-h`, `--help`	Print help for the current command path
`-V`, `--version`	Print CLI version
`-v`, `--verbose`	Print expanded diagnostics and source chains
`--profile`	Run this invocation against a specific profile
`--backend-url`	Run this invocation against a specific backend target

Output Modes

Human output is the default.
--json prints schema-validated command data to stdout.
--plain prints tab-separated rows to stdout when a command supports it.
Progress, warnings, validation errors, runtime errors, and structured errors are rendered on stderr.
--json and --plain are mutually exclusive.

Environment

PAW_HOME: Override the CLI state root for config and cache.
PAW_PROFILE: Set the active profile when --profile is not supplied.
PAW_BACKEND_URL: Override the backend target for this invocation.
XDG_CONFIG_HOME: Fallback config root when PAW_HOME is unset.
XDG_CACHE_HOME: Fallback cache root when PAW_HOME is unset.

Exit Codes

Code	Meaning
0	Success
1	Internal, local, or config error
2	Usage, validation, or ambiguous input source
5	Backend, network, external process, or dependency failure

Verification

bun run --filter @pawrrtal/cli typecheck
bun run --filter @pawrrtal/cli test
bun run --filter @pawrrtal/cli check
bun run skill-gen:check
just paw-cli-check

Pitfalls

-V prints the CLI version; -v enables verbose diagnostics.
PAW_HOME overrides both config and cache roots for the invocation.
paw context --json reports source labels such as flag, env:PAW_PROFILE, project:<path>, user:<path>, home-default, and unset.
paw context and paw whoami must not print secret values.
paw doctor is local-only in this slice and may warn when optional state has not been generated yet.

---name: pawdescription: Use when running the Bun-backed Paw CLI: local health, active context, shell completions, output modes, profiles, and verification of the supported command surface.---# pawUse `paw` when you need the supported Pawrrtal CLI surface. This CLI is a Bun-backed Effect v4 package at `packages/paw-cli`; do not route new work through the deleted Python CLI.## Commands| Command | Aliases | Summary || --- | --- | --- || `paw doctor` | - | Check local Paw CLI readiness || `paw context` | whoami | Print the active CLI context without secrets || `paw completions <shell>` | - | Generate shell completions |## Common Usage```bashjust paw --helpjust paw doctor --jsonjust paw context --jsonjust paw whoami --plainjust paw completions zshbun run --filter @pawrrtal/cli start -- doctor```## Global Options| Option | Meaning || --- | --- || `-h`, `--help` | Print help for the current command path || `-V`, `--version` | Print CLI version || `-v`, `--verbose` | Print expanded diagnostics and source chains || `--profile` | Run this invocation against a specific profile || `--backend-url` | Run this invocation against a specific backend target |## Output Modes- Human output is the default.- `--json` prints schema-validated command data to stdout.- `--plain` prints tab-separated rows to stdout when a command supports it.- Progress, warnings, validation errors, runtime errors, and structured errors are rendered on stderr.- `--json` and `--plain` are mutually exclusive.## Environment- `PAW_HOME`: Override the CLI state root for config and cache.- `PAW_PROFILE`: Set the active profile when --profile is not supplied.- `PAW_BACKEND_URL`: Override the backend target for this invocation.- `XDG_CONFIG_HOME`: Fallback config root when PAW_HOME is unset.- `XDG_CACHE_HOME`: Fallback cache root when PAW_HOME is unset.## Exit Codes| Code | Meaning || --- | --- || 0 | Success || 1 | Internal, local, or config error || 2 | Usage, validation, or ambiguous input source || 5 | Backend, network, external process, or dependency failure |## Verification```bashbun run --filter @pawrrtal/cli typecheckbun run --filter @pawrrtal/cli testbun run --filter @pawrrtal/cli checkbun run skill-gen:checkjust paw-cli-check```## Pitfalls- `-V` prints the CLI version; `-v` enables verbose diagnostics.- `PAW_HOME` overrides both config and cache roots for the invocation.- `paw context --json` reports source labels such as `flag`, `env:PAW_PROFILE`, `project:<path>`, `user:<path>`, `home-default`, and `unset`.- `paw context` and `paw whoami` must not print secret values.- `paw doctor` is local-only in this slice and may warn when optional state has not been generated yet.

Frequently asked questions

Paw