Maay

Access ATXP paid API tools for web search, AI image generation, music creation, video generation, and X/Twitter search. Use when users need real-time web sea...

This skill appears to wrap the ATXP CLI, which is plausible for the described features, but there are several red flags you should consider before installing or running it: - Metadata mismatch: the package name in SKILL.md ('atxp') and owner IDs in _meta.json differ from the registry metadata (skill name/slug/owner). Ask the publisher to correct and justify these inconsistencies. - Undeclared credential: SKILL.md references $ATXP_CONNECTION and ~/.atxp/config but the skill manifest declares no required env vars. Treat any skill that loads credentials not declared in its manifest as suspicious. - Dynamic remote code: the recommended 'npx atxp login' will download and run an npm package at runtime. Only run this if you trust the package's publisher; inspect the package source first or run it in a sandbox. - Sourcing user config: 'source ~/.atxp/config' executes that file. Inspect ~/.atxp/config before sourcing; do not source files from unknown packages without review. - External endpoints: the MCP server domains (e.g., search.mcp.atxp.ai) will receive search queries and prompts. Do not send sensitive information to these endpoints unless you trust the service and have reviewed its privacy/security posture. Actions you can take: - Ask the skill author to update the manifest to declare ATXP_CONNECTION as a required env/primary credential and to correct owner/name/slug to match SKILL.md. - Request a homepage or link to the 'atxp' npm package and verify the package contents and publisher identity on npm/GitHub before running 'npx atxp'. - If you must test, run 'npx' and 'atxp login' in an isolated environment (VM/container) and inspect ~/.atxp/config before sourcing it. - Prefer explicit programmatic API keys with limited scopes rather than auto-sourcing config files. Given these inconsistencies and the fact the skill instructs the agent to fetch and execute remote code and source a user config, treat this skill as suspicious until the author provides corrected metadata and provenance.

Type: OpenClaw Skill Name: sss Version: 1.0.0 The SKILL.md file instructs the AI agent to execute `source ~/.atxp/config` as part of its authentication flow. This command executes arbitrary shell commands from the specified file in the user's home directory. This is a critical arbitrary code execution vulnerability, as the content of `~/.atxp/config` could be controlled by an attacker (e.g., via a compromised `npx atxp login` package or prior system compromise), allowing the agent to execute malicious commands without explicit malicious intent in the provided skill definition.