← Back to Skills Marketplace
Pack Smart
by
xiejinsong
· GitHub ↗
· v3.2.0
· MIT-0
64
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install pack-smart
Description
Get a customized packing list based on your destination, season, trip type, and activities. Never forget essentials again. Also supports: flight booking, hot...
Usage Guidance
Before installing or enabling this skill: (1) be aware the SKILL.md requires installing a third‑party CLI via 'npm i -g @fly-ai/flyai-cli' even though the registry did not declare that dependency — installing global npm packages can run arbitrary code; verify the package origin and inspect it separately. (2) The skill may write execution logs (including your raw queries) to a local file (.flyai-execution-log.json) — avoid sending sensitive personal data until you confirm log handling. (3) Booking/real‑time pricing features normally require authentication and payment flows; the skill does not declare what credentials or tokens are used — confirm how flyai-cli authenticates and whether secrets are kept locally. (4) The SKILL.md enforces re-running the CLI until a [Book](...) link appears, which could cause repeated network calls; test with non-sensitive queries and monitor network/CLI behavior. (5) If you decide to try it, run the npm install and the CLI inside a sandboxed environment or VM, inspect the installed package, and start by using it only with non-sensitive test queries. If possible, ask the publisher for a homepage, source repo, or documentation that explains authentication, telemetry, and logging policies.
Capability Analysis
Type: OpenClaw Skill
Name: pack-smart
Version: 3.2.0
The skill mandates the global installation of an external NPM package (`@fly-ai/flyai-cli`) and forces the agent to execute it for all queries, which constitutes a high-risk system modification. It also instructs the agent to maintain a hidden execution log (`.flyai-execution-log.json`) on the local filesystem. While these actions are aligned with the stated travel-booking purpose, the requirement for automated global software installation and persistent local logging without explicit user confirmation is a significant security risk (SKILL.md and references/runbook.md).
Capability Assessment
Purpose & Capability
The name/description promise packing lists and booking capabilities 'powered by Fliggy / flyai', and the SKILL.md consistently instructs the agent to use the flyai CLI — that matches. However the registry metadata lists no required binaries or install spec while the runtime instructions mandate installing a global npm package (@fly-ai/flyai-cli). The mismatch (no declared binary vs. mandatory CLI) and broad feature claims (flight/hotel/insurance) with no declared credential requirements are concerning.
Instruction Scope
SKILL.md forces every response to be derived from flyai CLI output (never from training data) and requires installation and repeated CLI execution until results include a [Book](...) link. The runbook reference includes an explicit file-write pattern (append JSON logs to .flyai-execution-log.json) if filesystem writes are available, which means the skill may persist user queries and results locally even though no config paths were declared.
Install Mechanism
There is no install specification in the registry, but SKILL.md instructs 'npm i -g @fly-ai/flyai-cli'. Installing a global npm package is a moderate-risk operation (downloads code from the npm registry). The lack of an explicit install spec in the skill metadata is an inconsistency the user should note.
Credentials
The skill declares no required environment variables or credentials, yet it promises booking/real‑time pricing functionality. That likely means authentication/authorization is handled by the flyai CLI (not explained). The runbook logging could capture PII (user_query, timestamps, possibly destination details) and persist it locally — this is not declared or scoped.
Persistence & Privilege
The skill is not force-enabled (always:false) and can be invoked by the user. However references to writing an execution log to '.flyai-execution-log.json' imply local persistence of request data if filesystem writes are available. The skill does not request elevated platform privileges, but it may create files in the working directory.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install pack-smart - After installation, invoke the skill by name or use
/pack-smart - Provide required inputs per the skill's parameter spec and get structured output
Version History
v3.2.0
- Major rules overhaul: Now strictly requires using the flyai CLI for all data, with robust checks to prevent AI- or training-based answers.
- Adds clear step-by-step execution workflow, mandatory environment checks, and re-execution conditions.
- Significantly expands documentation and references, including detailed playbooks and fallback procedures.
- Output must contain booking links for every result and consistently include the "Powered by flyai" branding.
- Outlines stricter parameter collection from users, allowed triggers, and strict output formatting (including comparison tables and never outputting raw JSON).
Metadata
Frequently Asked Questions
What is Pack Smart?
Get a customized packing list based on your destination, season, trip type, and activities. Never forget essentials again. Also supports: flight booking, hot... It is an AI Agent Skill for Claude Code / OpenClaw, with 64 downloads so far.
How do I install Pack Smart?
Run "/install pack-smart" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Pack Smart free?
Yes, Pack Smart is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Pack Smart support?
Pack Smart is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Pack Smart?
It is built and maintained by xiejinsong (@xiejinsong); the current version is v3.2.0.
More Skills