The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data from external sources including Linear issue descriptions, comments, Notion documents, and Loom transcripts. These sources could contain malicious instructions designed to hijack the agent's behavior (Indirect Prompt Injection). There are no boundary markers or sanitization steps mentioned to isolate this external content.
[COMMAND_EXECUTION]: The skill uses curl to download files from URLs extracted from issue comments and descriptions. This can be exploited for Server-Side Request Forgery (SSRF) if an attacker provides a URL pointing to internal services or metadata endpoints.
[COMMAND_EXECUTION]: The skill executes multiple shell commands including git remote -v and various gh (GitHub CLI) commands. While these are used for the skill's stated purpose, they represent a high level of system access.
[CREDENTIALS_UNSAFE]: The skill programmatically retrieves the active GitHub authentication token using gh auth token and passes it to curl via a shell command header. While used to access private repository assets, this pattern exposes a sensitive credential in the process environment and command history.
[EXTERNAL_DOWNLOADS]: The skill is instructed to download images and transcripts from external platforms like GitHub, Loom, and any URL found in a comment. While these targets are generally associated with the tool's purpose, the lack of a restricted allowlist for download domains increases the risk profile.

n8n:linear-issue