Local AI email assistant that automates unread-mail processing, offline-safe classification and reply drafting, local draft editing/approval queues, calendar conflict checks, voice transcription, and confirmed SMTP sending when connectivity returns. Use when the user wants to manage email, check inbox, draft or edit replies, queue messages offline, approve/send email, handle meeting invites, or mentions email assistant, IMAP, SMTP, voice memo, local AI email.
82%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The required workflow uses `read_mailbox` to fetch unread emails via IMAP, which consumes untrusted outsider-authored email bodies and headers without prior selection.
The skill fetches instructions or code from an external URL at runtime, and the fetched content directly controls the agent’s prompts or executes code. This dynamic dependency allows the external source to modify the agent’s behavior without any changes to the skill itself.
The skill downloads ModelScope artifacts from https://modelscope.cn during automated setup or runtime model retrieval, which serves as an operational dependency with common/reputable unofficial or arbitrary third-party provenance.
anbeime/skill
cc0796c
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.