v1.0.1

Reveal Reviewer skills

ReviewClawScan verdict for this skill. Analyzed May 1, 2026, 5:52 AM.

Analysis

The skill is a coherent Reveal review helper, but it should be reviewed carefully because it can use your Reveal API key to create and submit persistent review records and screenshot links without a clear final confirmation step.

GuidanceInstall or use this skill only if you are comfortable letting the agent act on your Reveal reviewer account. Before any submission, ask the agent to show the selected task or self-review target, the draft findings, notes, and all screenshot URLs, then wait for your approval before sending them to Reveal.

Findings (3)

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

SeverityMediumConfidenceMediumStatusConcern

SKILL.md

"Fallback" ... "If no matching/aligned task exists, use Proactive Self-Review Mode" ... "POST `/self-reviews`" ... "Step 5: Submit the review" ... "POST `/submissions`"

The workflow authorizes the agent to switch into a write path and create or submit review records through the external API. The artifact shows the write operations but does not clearly require the user to approve the proactive self-review creation or final submission payload first.

User impactThe agent could create a proactive review or submit review content and screenshot links under your Reveal account when you expected only task discovery or draft feedback.

RecommendationRequire explicit user confirmation before creating a self-review, submitting a review, updating a self-review, or sending screenshot URLs. Show the target site, task ID, findings, notes, and screenshots before any POST/PATCH request.

Permission boundary

Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.

Identity and Privilege Abuse

SeverityLowConfidenceHighStatusNote

SKILL.md

"REVEAL_REVIEWER_API_KEY" environment variable ... "Authorization: Bearer $REVEAL_REVIEWER_API_KEY"

The skill uses a Reveal reviewer API key for authenticated calls. This is expected for the integration, but it gives the agent delegated authority over the user's reviewer account actions.

User impactAnyone or any agent workflow with access to this key may be able to act as the reviewer for supported Reveal API actions.

RecommendationUse a revocable, least-privileged Reveal reviewer key if available, keep it out of chat transcripts and logs, and rotate or revoke it when no longer needed.

Sensitive data protection

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

Insecure Inter-Agent Communication

SeverityLowConfidenceHighStatusNote

SKILL.md

"Take a screenshot using agent-browser's snapshot feature" ... "Screenshots should be URLs if your agent-browser supports image capture/upload"

The skill relies on another browser automation skill to capture screenshots and may submit uploaded screenshot URLs to Reveal. This is purpose-aligned, but the screenshot storage/upload boundary is not fully described in the artifact.

User impactScreenshots could capture sensitive page content and be linked in the Reveal submission if the user or agent navigates into authenticated or private areas.

RecommendationAvoid logging into personal accounts during reviews unless necessary, redact sensitive data, and have the agent list screenshot URLs for approval before submission.