Line Client

WarnAudited by ClawScan on May 10, 2026.

Overview

This is a coherent LINE client, but it asks the agent to use persistent LINE login tokens and broad account controls through code that was not included for review.

Install only if you are comfortable granting an agent broad access to your LINE account. Before using it, verify the referenced repository and helper code, protect or remove ~/.line-client/tokens.json when done, and require explicit confirmation for every message, contact change, or group-management action.

Findings (5)

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

If installed and used, the agent may be able to read and act as the user in LINE until the token expires or is revoked.

Why it was flagged

The skill creates and uses persistent LINE session credentials, including refresh tokens, which can grant delegated access to the user's LINE account.

Skill content
- **Token storage:** `~/.line-client/tokens.json` ... `result.auth_token, result.mid, result.refresh_token`
Recommendation

Only use this if you trust the implementation, keep the token file protected, revoke/re-login if compromised, and require explicit user approval before any account-changing action.

What this means

A mistaken or over-autonomous agent action could send messages, delete/unsend messages, add or block contacts, create groups, invite people, or remove members from chats.

Why it was flagged

The documented methods allow broad messaging, contact, and group mutations, but the visible skill instructions do not define explicit confirmation, scoping, or safety checks for these high-impact operations.

Skill content
`send_message(to, text, ...)`, `unsend_message(message_id)`, `block_contact(mid)`, `create_chat(name, target_mids)`, `invite_into_chat(chat_id, mids)`, `delete_other_from_chat(chat_id, mids)`
Recommendation

Require confirmation with exact recipient, message text, and intended action before any send, delete, contact, or group-management operation.

What this means

The actual code that would handle LINE tokens and account actions is outside the reviewed package, increasing the risk of unexpected behavior or supply-chain changes.

Why it was flagged

The skill depends on external/local executable components that are not included in the provided artifact set, so their behavior, dependencies, and token handling could not be reviewed.

Skill content
- **Repo:** `/data/workspace/line-client` ([github.com/2manslkh/line-api](https://github.com/2manslkh/line-api))
- **Main client:** `src/chrome_client.py`
- **HMAC signer:** `src/hmac/signer.js`
- **WASM files:** `lstm.wasm` + `lstmSandbox.js`
Recommendation

Review and pin the referenced repository and helper files before use, and prefer a packaged skill that includes the reviewed implementation and install specification.

What this means

A local helper process may remain active while signing requests for the LINE client.

Why it was flagged

The skill documents an auto-starting local signing helper. This may be normal integration plumbing, but the visible artifact does not describe its binding, authentication, shutdown, or cleanup behavior.

Skill content
- **HMAC signer:** `src/hmac/signer.js` (Node.js, auto-starts on port 18944)
Recommendation

Verify that the signer binds only to a local interface, cannot be accessed by other users or sites, and is stopped when no longer needed.

What this means

LINE messages, contacts, group information, and profile data may be requested through the documented gateway as part of normal use.

Why it was flagged

The skill routes LINE client operations through an external gateway. This is disclosed and aligned with the stated purpose, but it involves sensitive message and contact data.

Skill content
Provides HMAC-signed API access through the Chrome extension gateway (line-chrome-gw.line-apps.com).
Recommendation

Confirm that the gateway domain and account flow are legitimate for your use case before authenticating or sending sensitive messages.