Vocabulary Builder

PassAudited by ClawScan on May 1, 2026.

Overview

This is a coherent vocabulary-tracking skill that stores learning progress in a local file and may support scheduled quizzes, with no evidence of hidden code, credential use, or data exfiltration.

This skill appears safe to install if you want a local vocabulary tracker. Be aware that it will save words and learning history in `memory/vocabulary.md` by default, and it may participate in scheduled quiz reminders if your environment supports heartbeat or cron triggers.

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.

What this means

Words and practice records may be saved automatically unless the user explicitly asks not to save them.

Why it was flagged

The skill directs the agent to modify the local vocabulary tracker by default after a user provides a word. This is expected for a vocabulary tracker, but it is still a user-impacting file write.

Skill content
“Add to tracker immediately unless the user says not to add it”
Recommendation

Install only if you want automatic vocabulary tracking, and tell the agent not to add a word when you want a one-off explanation.

What this means

Your reading context and learning history can persist across sessions and influence future quizzes or progress tracking.

Why it was flagged

The skill maintains persistent memory containing words, book/page references, context sentences, and practice history. This is central to spaced repetition, but users should understand that the information is retained and reused.

Skill content
“Vocabulary tracker: `memory/vocabulary.md` — all word data lives here”
Recommendation

Avoid storing private excerpts or sensitive personal context in the tracker, and periodically review or edit `memory/vocabulary.md`.

NoteMedium Confidence
ASI10: Rogue Agents
What this means

If the host environment uses scheduled triggers, the agent may initiate vocabulary quizzes without a fresh manual request each time.

Why it was flagged

The skill contemplates scheduled quiz activity triggered by a heartbeat. The visible instructions include anti-spam limits, so this appears purpose-aligned rather than rogue, but it is still autonomous reminder behavior users should notice.

Skill content
“or when a cron/heartbeat triggers a scheduled quiz”
Recommendation

Use scheduled quiz behavior only if you want reminders, and check the host’s scheduling or heartbeat settings if quizzes become unwanted.