Security audit
Security checks for vulnerabilities and agentic risk
This skill is a coherent Beeper messaging integration, but it asks for broad chat access and persistent token storage while giving weak guardrails for sending or changing messages.
Review this carefully before installing if your Beeper account includes sensitive personal or work chats. Use the minimum accounts you need, avoid storing the token in plaintext when possible, revoke or rotate the token if you stop using the skill, and require the agent to confirm before it sends, reacts, archives, marks read, downloads attachments, or creates chats/reminders.
The skill markets cross-platform search, inbox summarization, attachment access, and contact/message retrieval without prominently warning that it exposes highly sensitive communications and social graph data. Users may authorize or invoke the skill without understanding that it can access private content across multiple services from a single interface.
The skill instructs users to generate a long-lived Beeper access token for automation, which grants broad access to unified messaging data and actions across many platforms. If mishandled, leaked, or reused by other local tools, the token could enable reading messages, contacts, attachments, and sending or modifying communications.
### 3. (Optional) Add Your Beeper Token
For smoother automation, grab an access token:
1. Beeper Desktop → Settings → Developers
2. Click "Create Access Token"
The documentation tells users to place the access token in a local configuration file, creating persistent plaintext credential storage for a highly sensitive messaging integration. This raises the risk of token theft via local compromise, backups, accidental sharing, or other tools that read the config file.
For smoother automation, grab an access token:
~/.clawdbot/clawdbot.json:The skill advertises state-changing capabilities such as sending messages, reacting, setting reminders, and marking chats as read, but it does not include an explicit caution that these actions modify user communications and account state. That omission increases the risk of accidental or unauthorized actions, especially when driven by natural-language requests.
Storing the Beeper access token in ~/.clawdbot/clawdbot.json creates session persistence beyond the immediate interactive flow. Persistent credentials increase the blast radius of any local compromise and make unintended future access to messaging accounts more likely.
For smoother automation, grab an access token:
~/.clawdbot/clawdbot.json:The skill encourages broad natural-language commands for searching, summarizing, and sending messages across many platforms without requiring clear confirmation boundaries. In a high-impact messaging context, ambiguous prompts can cause unintended disclosure of private message history or accidental outbound communications on the user's behalf.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
See all connected platforms in your Beeper:
curl -s "${BEEPER_API_URL}/v1/accounts" \
-H "Authorization: Bearer ${BEEPER_ACCESS_TOKEN}"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
-H "Authorization: Bearer ${BEEPER_ACCESS_TOKEN}" | jq -r '.[0].id')
# React with thumbs up
curl -X POST "${BEEPER_API_URL}/v1/chats/{chatID}/messages/${LAST_MSG}/reactions" \
-H "Authorization: Bearer ${BEEPER_ACCESS_TOKEN}" \
-H "Content-Type: application/json" \
-d '{"emoji": "👍"}'
No suspicious patterns detected.