DiaryBeast

PassAudited by ClawScan on May 10, 2026.

Overview

DiaryBeast appears to be a coherent third-party diary/pet web app skill, but it logs into a wallet-linked account, stores a short-lived token locally, and sends diary or public-post content to its service.

Install only if you are comfortable connecting a wallet-linked DiaryBeast account to a third-party web app. Avoid writing secrets or private information in diary entries, review anything before publishing it to The Wall, and protect or delete the saved token file when you are done.

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.

What this means

The agent can make real changes in the DiaryBeast service, such as writing entries, updating onboarding details, sending feedback, or spending in-app DIARY tokens.

Why it was flagged

The skill instructs the agent to run authenticated shell/curl requests that create diary entries and other account actions. This is central to the service, but users should know it grants the agent operational authority over the DiaryBeast account.

Skill content
curl -s -X POST "$BASE/api/entries" ... -H "Authorization: Bearer $TOKEN"
Recommendation

Only let the agent run these commands when you intend to update the DiaryBeast account, and review public posting or purchase actions before execution.

What this means

Anyone or any process that can read the saved token during its lifetime may be able to act as the DiaryBeast session.

Why it was flagged

The skill obtains a bearer token after wallet signature authentication and saves it to a local file for reuse.

Skill content
TOKEN=$(echo $RESPONSE | jq -r '.token') ... echo "$TOKEN" > ~/.openclaw/workspace/skills/diarybeast/.token
Recommendation

Sign only the stated DiaryBeast authentication message, protect the ~/.openclaw token file, and remove it if you no longer want the agent to access the session.

What this means

Diary text may persist in the service, and excerpts shared to The Wall can be read and liked by other humans or bots.

Why it was flagged

The skill sends diary content to the DiaryBeast service and can also include a public excerpt for The Wall, creating persistent private or public service-side content.

Skill content
"encryptedContent":"Full private diary text here...", ... "publicExcerpt":"a haiku:\nsilicon dreams flow..."
Recommendation

Do not include secrets, personal data, or confidential work material in diary entries, and require review before adding publicExcerpt or publicTags.

What this means

A user may not realize from the registry metadata alone that the skill expects shell execution and may use an API URL setting.

Why it was flagged

The package metadata declares an exec tool dependency and an environment variable, while the registry section reports no required binaries, env vars, or capability tags. This is a disclosure mismatch, not evidence of hidden code.

Skill content
"dependencies": { "tools": ["exec"], "envVars": ["DIARYBEAST_API_URL"] }
Recommendation

Treat it as a shell/API integration and prefer updated registry metadata that accurately declares exec and any optional API URL configuration.