Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Competitive Intelligence
v1.0.0Performs competitive intelligence by researching competitors and analyzing their online presence using web search and site scraping. Use when the user mentio...
⭐ 0· 105·0 current·0 all-time
byMario Karras@mariokarras
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill's name/description (competitive intelligence via web search and scraping) aligns with the runtime instructions. However, the SKILL.md explicitly calls local CLIs (node tools/clis/exa.js and node tools/clis/firecrawl.js) and references product-marketing-context files; these dependencies are not listed in the skill's required binaries or config paths, which is inconsistent.
Instruction Scope
The instructions tell the agent to read local files (.agents/product-marketing-context.md or .claude/product-marketing-context.md) and to execute local Node scripts under tools/clis. Reading arbitrary workspace files and running local scripts are actions beyond simple web queries and can expose sensitive local context or run unreviewed code. The SKILL.md does not limit which files to read beyond the two named context files, nor does it declare those paths in the metadata.
Install Mechanism
No install spec (instruction-only), which minimizes disk-write risk. But the skill presumes the presence of specific local CLIs; because nothing is installed or declared, those commands could fail or, if present, run arbitrary local code. This implicit dependency is a risk if the provenance of the local tools is unknown.
Credentials
The skill requests no environment variables, credentials, or config paths in metadata. That is proportionate to the stated purpose. Note: the SKILL.md still instructs reading local files not declared in requires.config, which is an access-not-declared issue (covered under instruction_scope).
Persistence & Privilege
The skill is not 'always' enabled and is user-invocable. It does not request persistent privileges or modify other skills. Autonomous invocation is allowed (platform default) but not combined with other high-risk flags here.
What to consider before installing
Before installing or enabling this skill, consider the following:
- Verify local tool provenance: The skill's runtime commands call node tools/clis/exa.js and firecrawl.js. Confirm those scripts exist, inspect their source, and ensure they are from a trusted origin before allowing the agent to execute them.
- Confirm workspace file access is acceptable: The skill instructs the agent to read .agents/product-marketing-context.md (or .claude/...). Decide whether the agent should read those files or whether you will supply the context manually to avoid leaking unrelated workspace data.
- Prefer explicit declarations: Ask the skill author (or maintainers) to list required binaries and config paths in metadata so you can review them up front.
- Sandbox and least privilege: If you proceed, run the skill in a sandboxed environment or with restricted agent permissions until you confirm behavior. Limit autonomous runs if you don't want the agent to execute local scripts without supervision.
- Scraping/legal considerations: The skill recommendations include website scraping. Make sure scraping targets and frequency comply with site terms of service and legal constraints.
If the author updates the metadata to declare the required CLIs/config paths and provides the source for the referenced tools (or bundles safe, audited implementations), I would raise my confidence and the assessment could become benign.Like a lobster shell, security has layers — review code before you run it.
latestvk97191ryytympaszawh6px3dzh834k1d
License
MIT-0
Free to use, modify, and redistribute. No attribution required.