v1.0.0

reMarkable Tablet Sync

ReviewClawScan verdict for this skill. Analyzed May 1, 2026, 5:26 AM.

Analysis

The skill is purpose-aligned, but it deserves review because it uses a persistent reMarkable cloud token, can bulk transfer or upload tablet content, and may store private journal contents in agent memory.

GuidanceInstall only if you are comfortable giving the agent ongoing access to your reMarkable cloud content. Prefer a dedicated sync folder, verify the rmapi download and Python packages, review file lists before downloads/uploads, and do not let sensitive journals be added to memory unless you understand how that memory can be reviewed and deleted.

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

SeverityMediumConfidenceHighStatusConcern

SKILL.md

# Bulk upload
rmapi mput ./local-folder/ "Remote Folder/"

# Create folder on tablet
rmapi mkdir "New Folder"

The skill documents raw cloud-sync commands that can bulk upload local content and mutate the tablet account, but the artifact does not require review or confirmation before these actions.

User impactA mistaken or overly broad instruction could upload unwanted files or create unwanted content in the user's reMarkable cloud/tablet environment.

RecommendationUse an explicit allowlist of target folders, show planned uploads before running them, and require user confirmation for bulk or account-mutating commands.

Agentic Supply Chain Vulnerabilities

SeverityLowConfidenceHighStatusNote

SKILL.md

curl -L https://github.com/ddvk/rmapi/releases/latest/download/rmapi-linux-amd64 -o ~/bin/rmapi
chmod +x ~/bin/rmapi

The setup instructs users to download and execute the latest external binary rather than a pinned or verified release; this is user-directed and purpose-aligned but still a provenance risk.

User impactIf the downloaded binary or package source is compromised or changes unexpectedly, the local tool used with the reMarkable account could be unsafe.

RecommendationVerify the rmapi source, pin versions where possible, check release hashes/signatures if available, and install Python dependencies from trusted sources.

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

SeverityMediumConfidenceHighStatusConcern

SKILL.md

Token saved to `~/.rmapi` — future runs are automatic

This shows the skill relies on a persistent reMarkable Cloud session/profile; after one-time login, later rmapi commands can act on the tablet account without another authentication prompt.

User impactAn agent using this skill may continue accessing the user's reMarkable cloud account after setup, including private notes and upload actions.

RecommendationDeclare this credential requirement clearly, limit use to a dedicated sync folder or account where possible, and require explicit confirmation before account-changing actions.

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

SeverityMediumConfidenceHighStatusConcern

SKILL.md

**Journal entries** — Fetch handwritten thoughts → interpret → append to memory/journal

The workflow contemplates converting private handwritten journal content into persistent memory/journal state, without retention, deletion, consent, or reuse boundaries.

User impactSensitive personal notes could persist in the agent's memory and influence future tasks beyond the original sync request.

RecommendationOnly store extracted journal content with explicit user approval, define retention/deletion behavior, and avoid adding sensitive content to shared or long-lived memory by default.