Back to skill
v1.0.0

Youtube Learning

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 8:36 AM.

Analysis

This is an instruction-only YouTube learning skill whose described behavior is coherent, but users should pay attention to optional note export, calendar scheduling, and saved learning-history features.

GuidanceThis skill appears safe to install as an instruction-only educational helper. Before using optional integrations, decide where notes should be saved, whether exports to Notion or Obsidian are allowed, whether calendar events may be created, and how long learning history or progress data should be kept.

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
SeverityLowConfidenceHighStatusNote
SKILL.md
| Transcript API | Fetch YouTube transcripts |\n| File System | Save notes locally |\n| Note Apps | Export to Notion, Obsidian |\n| Calendar | Schedule learning sessions |

The skill describes using external transcript tools, local file writes, note-app exports, and calendar scheduling. These are aligned with the learning purpose, but they can create or modify user data and should be scoped by the user.

User impactIf enabled, the agent may save files, export notes, or create calendar items related to learning sessions.
RecommendationConfirm each save, export, or calendar action; specify the destination folder, app, workspace, and schedule before letting the agent act.
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
SeverityLowConfidenceMediumStatusNote
_meta.json
"integrations": [\n    "YouTube transcript API",\n    "File system",\n    "Notion",\n    "Obsidian"\n  ]

The metadata lists third-party and local integrations. No credentials are declared or hardcoded, so this is not suspicious by itself, but Notion or similar exports may require account permissions if connected.

User impactGranting note-app access could let the agent create or modify content in connected workspaces.
RecommendationUse least-privilege app connections, avoid broad workspace permissions where possible, and review generated notes before publishing or sharing them.
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
- Progress tracking\n...\n- Save videos to collections\n- Tag and categorize\n- Track watched status\n- Resume where you left off

The skill includes persistent learning-state features. This is expected for a learning tool, but the artifacts do not specify storage location, retention, or how saved notes and progress are reused.

User impactSaved collections, progress, and notes may reveal learning interests and could influence future summaries or recommendations.
RecommendationKeep saved learning data in a known location, review or delete it when no longer needed, and treat generated notes as study aids rather than authoritative memory.