Competitive Intel
Analysis
This skill fits competitive research, but it under-discloses that it needs a Bright Data login, paid CLI activity, and an unpinned remote installer.
Findings (9)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
**Never answer competitive questions from training knowledge alone.** Always gather live data first using `bdata` commands
This explicitly forces the agent to prefer external tool use over answering from existing context. It is purpose-aligned for real-time intelligence, but it changes the normal response path.
Multi-competitor deep dive (3+) | 15-30 calls | ~$0.10-$0.50
The skill expects multiple billable bdata tool calls, and the provided capability signals include can-make-purchases. The artifacts give cost guidance but do not require user approval before cost-bearing calls.
curl -fsSL https://cli.brightdata.com/install.sh | bash
The skill instructs users to execute a remote installer script directly, while the package has no install spec or pinned dependency provenance.
curl -fsSL https://cli.brightdata.com/install.sh | bash
This is an install-to-run flow that executes downloaded shell code, which is more powerful than an instruction-only skill's declared installation footprint.
**Parallelize** — Run independent `bdata` calls in parallel using multiple Bash tool calls in a single response
The skill encourages parallel external calls, so an incorrect target or overly broad request could be amplified into several scraping/search actions. The cost-efficiency guidance mitigates this but is not a hard control.
That's it. No env vars, no zone config, no API keys to manage.
This wording downplays setup and credential complexity even though the same prerequisite section requires bdata login and the workflow uses cost-bearing external calls.
Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.
One-time login completed: `bdata login`
The skill requires an authenticated Bright Data account session even though the registry metadata declares no primary credential or required environment variables.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
Always gather live data first using `bdata` commands, then analyze and synthesize.
The skill intentionally places external scraped web content into the agent's working context and uses it as evidence for analysis. This is purpose-aligned, but such content can be manipulated by website owners or reviewers.
Real-time competitive intelligence powered by live web data. Combines Bright Data CLI (`bdata`) for data collection
The skill routes research queries, target URLs, and collection tasks through an external provider CLI. This is disclosed and purpose-aligned, but the data boundary depends on Bright Data.