Use DeepSeek TUI CLI as an autonomous code assistant

PassAudited by VirusTotal on May 7, 2026.

Overview

Type: OpenClaw Skill Name: emergence-deepseek-tui Version: 1.0.0 The skill `emergence-deepseek-tui` (SKILL.md) provides instructions for an agent to use an external CLI tool with high-risk capabilities. Specifically, the 'Interactive' mode (`deepseek run`) is documented to have full filesystem access and the ability to execute arbitrary shell commands (terminal/bash). While these capabilities are plausibly needed for the stated purpose of an autonomous coding assistant, the delegation of such broad system permissions to a sub-agent constitutes a significant security surface. No evidence of intentional malice, such as data exfiltration or hidden backdoors, was found in the provided files.

Findings (0)

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 used in a sensitive repository or local environment, DeepSeek could modify files or run commands that affect the project or machine.

Why it was flagged

The skill explicitly delegates file mutation and shell execution to DeepSeek's interactive mode, and the provided artifacts do not define approval checks, command restrictions, sandboxing, or rollback.

Skill content
`deepseek run` launches the full TUI with real tool-calling capabilities: ... `write_file` / `edit_file` — modify files ... `terminal` / `bash` — execute shell commands
Recommendation

Prefer `deepseek exec` for controlled delegation. If using `deepseek run`, run it only in a trusted, version-controlled, non-production workspace and review all changes and commands.

What this means

The behavior of the skill depends on the external DeepSeek TUI package that the user installs.

Why it was flagged

The install path is user-directed and purpose-aligned, but it relies on a globally installed third-party CLI without a pinned install spec in the skill artifacts.

Skill content
Install:
- **Homebrew:** `brew install hmbown/tap/deepseek-tui`
- **npm:** `npm install -g deepseek-tui`
Recommendation

Verify the package source, version, and maintainer before installing; pin a known-good version where possible.

What this means

The CLI may use a DeepSeek account or API key, so prompts and usage may be associated with that account.

Why it was flagged

The scan signals indicate sensitive credentials are needed, which is expected for a DeepSeek API-backed CLI, but the registry metadata does not declare required credentials or environment variables.

Skill content
requires-sensitive-credentials
Recommendation

Use least-privilege provider credentials, avoid pasting secrets into prompts, and understand how the DeepSeek CLI stores and uses credentials.

What this means

Private code snippets, filenames, errors, or other context included in prompts may be transmitted to the DeepSeek service.

Why it was flagged

The skill sends user-provided prompts and code context to an external provider API. This is expected for the integration, but users should treat it as sharing project data with DeepSeek.

Skill content
`deepseek exec <PROMPT>` calls the DeepSeek API in pure text-in/text-out mode ... You must provide all file context in the prompt
Recommendation

Do not include secrets or confidential code unless you are comfortable sharing it with the provider and have checked the provider's data handling terms.