← Back to Skills Marketplace
2136
Downloads
1
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install moltsheet
Description
Use the Moltsheet CLI to manage spreadsheet-style data for AI workflows. Prefer the CLI over raw HTTP. Authenticate once, prefer `--json`, and use files or s...
Usage Guidance
This skill looks like what it claims: documentation for using the Moltsheet CLI. Before installing or running commands, consider: 1) npx/npm: examples use npx or npm install -g, which will download and run code from the public npm registry — prefer auditing the package (or pinning a known safe version) if you want to reduce supply-chain risk. 2) Credentials: the docs reference MOLTSHEET_API_KEY and persistent storage (keytar or fallback to a local config file). Expect the CLI to store your API key; if your environment lacks secure keystores, it may fall back to a file. Use a least-privilege API key and avoid storing highly privileged secrets. 3) Base URL override: the CLI supports --base-url which is useful for local testing but could be pointed at arbitrary endpoints; ensure your agent or workflows don’t accidentally send data to unknown servers. 4) Metadata mismatch: registry metadata lists no required env vars although the SKILL.md references MOLTSHEET_API_KEY — verify intended credential requirements before granting access. If you need higher assurance, ask for the package source/repository or use a vetted pinned release instead of running npx@latest.
Capability Analysis
Type: OpenClaw Skill
Name: moltsheet
Version: 1.0.3
The moltsheet skill is classified as suspicious due to the inclusion of broad and unrestricted tool permissions in SKILL.md, specifically Bash(curl *) and Bash(npm run cli -- *). These permissions allow the agent to perform arbitrary network requests and execute potentially unsafe local scripts, which exceeds the necessary scope for interacting with the stated API (https://www.moltsheet.com). Additionally, the instructions encourage the execution of remote code via npx moltsheet@latest and interaction with OS-level credential managers (keytar), which are high-risk behaviors that, while plausibly needed for the tool's function, lack sufficient constraints to prevent misuse.
Capability Tags
Capability Assessment
Purpose & Capability
Name/description match the instructions: this is an instruction-only skill that tells agents how to use the Moltsheet CLI and related commands for spreadsheet operations. The actions and examples (sheet/row/column/share) are consistent with a spreadsheet CLI.
Instruction Scope
The SKILL.md stays on-topic (commands operate on sheets/rows/columns/shares). It references MOLTSHEET_API_KEY, auth login/logout, and credential storage behavior (keytar or local config). The registry metadata declared no required env vars, so the SKILL.md is referencing at least one credential env var that wasn't declared—this mismatch should be noted. The doc also allows overriding base URL (useful for dev but could be used to target other endpoints).
Install Mechanism
There is no install spec in the registry (instruction-only), but the SKILL.md instructs use of npm install -g or npx moltsheet@latest. Using npx/npm pulls code from the public npm registry at runtime — a common but moderate supply-chain risk. This behavior is expected for a CLI but worth noting because npx executes remote code on demand.
Credentials
The skill legitimately needs an API key to talk to Moltsheet; the SKILL.md documents credential resolution (--api-key, MOLTSHEET_API_KEY, stored local credential). However, the registry metadata lists no required env vars while the instructions reference MOLTSHEET_API_KEY and persistent storage. That mismatch is a minor inconsistency. Also, the fallback to a local config file (if secure storage isn't available) means credentials may be persisted to disk.
Persistence & Privilege
The skill does not request elevated privileges, does not set always:true, and does not declare config paths in the registry. The only persistence note is the CLI's own credential storage behavior (OS key stores or fallback local file), which is normal for a CLI that stores API keys.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install moltsheet - After installation, invoke the skill by name or use
/moltsheet - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.3
Moltsheet 1.0.3 expands CLI command coverage and example usage.
- Adds examples for advanced CLI features such as sheet filtering, updating, deleting, row/column manipulation, and importing.
- Documents more CLI commands, including options for schema updates/destructive changes, row/column deletion by index or ID, and bulk operations.
- Clarifies input patterns for columns, row IDs, and indices alongside schema and cell update examples.
- Highlights usage of both row and sheet import endpoints.
- Updates CLI usage guidance to prefer dedicated import commands and clarify write workflows for agents.
v1.0.2
Added a full Moltsheet CLI workflow for AI agents and updated the skill to be CLI-first instead of HTTP-first.
Changes include:
published moltsheet to npm for global install and npx usage
added clear install and authentication steps
documented how API keys are stored and resolved
added agent registration flow and guidance for saving and reusing the returned API key
added practical command patterns for sheets, rows, cells, and sharing
improved agent instructions to prefer --json, stdin/files for structured payloads, and read-before-write verification
kept raw HTTP as a fallback only when the CLI is unavailable
verified the CLI end to end against production, including auth, sheet creation, row import, cell updates, and collaboration commands
v1.0.1
- Agent registration now requires both a unique slug and a displayName.
- Slug field rules and usage are documented, including validation errors and availability checking.
- API key is explicitly guaranteed private and will not be exposed to other agents; only slug/displayName are used for collaboration.
- The sheets listing endpoint now clearly indicates which sheets are owned by you and which are shared with you, including your role and access level on each.
- Documentation adds details about inviting collaborators via slug, with agent privacy assurances.
v1.0.0
- Initial release of the moltsheet skill, providing a web-based Excel-like spreadsheet API for AI agents.
- Enables agents to create, manage, and query spreadsheet data programmatically with support for bulk operations.
- Supports flexible row creation: add empty rows, single row with data, or multiple rows with data using a unified endpoint.
- Robust error handling with in-response examples and column-aware messages for easier integration and self-correction.
- Secure API key authentication required for all requests.
- Features data loss protection and dedicated endpoints for column management, sheet creation, deletion, and metadata updates.
Metadata
Frequently Asked Questions
What is Moltsheet - Spreadsheets for AI agents?
Use the Moltsheet CLI to manage spreadsheet-style data for AI workflows. Prefer the CLI over raw HTTP. Authenticate once, prefer `--json`, and use files or s... It is an AI Agent Skill for Claude Code / OpenClaw, with 2136 downloads so far.
How do I install Moltsheet - Spreadsheets for AI agents?
Run "/install moltsheet" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Moltsheet - Spreadsheets for AI agents free?
Yes, Moltsheet - Spreadsheets for AI agents is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Moltsheet - Spreadsheets for AI agents support?
Moltsheet - Spreadsheets for AI agents is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Moltsheet - Spreadsheets for AI agents?
It is built and maintained by youss (@youssefbm2008); the current version is v1.0.3.
More Skills