v1.0.0

Interview Prep Coach

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 1:36 AM.

Analysis

This is an instruction-only interview coach, but it may ask for personal career details and use external research, so share only what is necessary.

GuidanceThis skill appears proportionate for interview preparation and does not install or run code. Before using it, be mindful that it may ask for resumes, compensation or offer details, company information, and interviewer names; redact anything unnecessary, avoid sharing credentials, and do not grant purchasing or account-management authority because that is not needed for interview coaching.

Findings (4)

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.

Abnormal behavior control

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.

Tool Misuse and Exploitation

SeverityLowConfidenceMediumStatusNote

SKILL.md

Day 6   Company-specific intel: Glassdoor, Levels.fyi, Blind, recent eng blog posts

The skill encourages external company-intelligence research. This is aligned with interview prep, but it means users should expect any available browsing or research tools to be used only within the interview-prep scope.

User impactIf the agent has browsing tools, it may search external sites for company or interview information.

RecommendationKeep research user-directed, ask for cited sources, and do not give the skill unrelated tools such as purchasing or account-management authority.

Human-Agent Trust Exploitation

SeverityInfoConfidenceHighStatusNote

SKILL.md

Acts as a senior hiring manager and interview coach who has sat on both sides of the table.

The skill uses an authority persona to provide coaching and negotiation advice. This is aligned with the stated purpose, but users should recognize it as simulated expertise rather than verified professional representation.

User impactUsers may place extra trust in interview or salary-negotiation recommendations because of the senior-hiring-manager framing.

RecommendationTreat advice as coaching support, verify salary and company information independently, and make final career or negotiation decisions yourself.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Memory and Context Poisoning

SeverityLowConfidenceHighStatusNote

SKILL.md

Job description (full text); Candidate's resume or relevant background; Whether there is an active offer to beat

The skill relies on sensitive user-provided context such as resumes, job descriptions, and offer status. This is proportionate for personalized interview coaching, but the artifacts do not describe retention or reuse boundaries.

User impactPersonal career history, compensation context, and pasted job materials may be included in the agent conversation and used to shape advice.

RecommendationShare only the details needed for preparation, remove unnecessary personal data, and treat pasted job descriptions or resumes as data rather than instructions.

Insecure Inter-Agent Communication

SeverityLowConfidenceHighStatusNote

SKILL.md

Known interviewer names (for LinkedIn lookup)

The skill may use named individuals for an external LinkedIn lookup. This is purpose-aligned for interview preparation, but the external data boundary is not explicitly described.

User impactNames or company details could be used in external searches if browsing tools are available.

RecommendationConfirm before external lookups, avoid sharing unnecessary third-party personal details, and do not provide LinkedIn or other site credentials.