slack-user-cli

Pass

Audited by Gen Agent Trust Hub on Nov 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The README provides instructions to install the uv package manager by downloading a shell script from its official domain (astral.sh) and piping it to a shell. This is a standard installation method for a well-known development tool.
  • [COMMAND_EXECUTION]: The main implementation script utilizes subprocess.run to call the macOS pbpaste utility for importing workspace configuration from the clipboard and invokes an external claudine-secret helper to manage credentials via a secure vault integration.
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves reading and searching Slack message history, which creates a surface for indirect prompt injection from untrusted external content. To mitigate this risk, the instructions explicitly require the agent to seek user approval via AskUserQuestion before performing any visible write actions, such as sending messages or clicking buttons, in shared channels.
  • [DATA_EXFILTRATION]: The tool extracts Slack session credentials, including xoxc- tokens and d cookies, from the local Slack application's data stores. This behavior is documented as the tool's intended mechanism for authentication. The script implementation includes basename-sanitization for file downloads to prevent path traversal and ensures that sensitive local configuration files are created with restricted file permissions (0o600).
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 3, 2026, 03:40 PM