Second Brain: Turn conversations into lasting knowledge

ReviewAudited by ClawScan on May 1, 2026.

Overview

This skill is a coherent personal knowledge-base integration that stores and retrieves user-confirmed notes through Ensue, with sensitive behavior disclosed and aligned to its purpose.

This appears safe to install if you intend to use Ensue as a persistent knowledge base. Before using it, make sure you trust the Ensue service with the notes you save, provide only the intended API key, and follow the skill’s own guidance to review drafts and avoid credentials, private paths, tokens, or other secrets.

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.

Low

#ASI03: Identity and Privilege Abuse

What this means

Anyone installing the skill must provide an Ensue API key, which grants the skill access to the user’s Ensue knowledge base according to that key’s permissions.

Why it was flagged

The wrapper uses an Ensue API key to authenticate calls to the external service; this is expected for the integration and is also disclosed in the skill metadata.

Skill content

-H "Authorization: Bearer $ENSUE_API_KEY"

Recommendation

Use a dedicated Ensue API key if possible, rotate it if exposed, and avoid saving secrets or highly sensitive personal data.

Medium

#ASI06: Memory and Context Poisoning

What this means

Saved notes may become future context for the assistant, so inaccurate, stale, or overly private content could affect later answers.

Why it was flagged

The skill stores persistent knowledge entries and enables embedding/search behavior, meaning saved content can be retrieved and reused in later tasks.

Skill content

Create entry: ... create_memory ... "embed":true

Recommendation

Review drafts before saving, keep entries accurate and non-sensitive, and periodically prune or update old memories.

Low

#ASI02: Tool Misuse and Exploitation

What this means

The assistant can mutate the user’s knowledge base when directed, but the documented workflow is user-controlled and purpose-aligned.

Why it was flagged

The skill can create and update remote knowledge entries, but the instructions add a user-confirmation workflow before saving.

Skill content

Always confirm before saving: ... Show draft of what will be saved ... Save after confirmation

Recommendation

Confirm saves or updates only after reviewing the draft and intended namespace/path.