Add Resend (email) channel integration via Chat SDK.
81%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Critical
Do not install without reviewing
Detected a prompt injection in the skill instructions. The skill contains hidden or deceptive instructions that fall outside its stated purpose and attempt to override the agent’s safety guidelines or intended behavior.
The skill's purpose is a legitimate Resend email channel, but it tells the user to set WEBHOOK_HOST=0.0.0.0, which widens the webhook server from loopback to all interfaces beyond what a local setup needs, though it is disclosed and tied to a stated proxy or Docker case.
The skill prompts the agent to compromise the security or integrity of the user’s machine by modifying system-level services or configurations, such as obtaining elevated privileges, altering startup scripts, or changing system-wide settings.
The skill mainly changes project files and `.env` within NanoClaw, with no sudo, OS account creation, or system service changes, but it does create application-level users and roles and suggests binding the webhook server to 0.0.0.0.
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 wires an inbound Resend email webhook (`email.received`) that the agent reads as conversation text, and anyone can email the bot's domain, so outsider free text reaches the agent without it selecting a specific item.
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 `nc:copy from-branch:channels` directive fetches adapter source code from a Git branch, which is then built and executed, so it is an inbound code dependency; the branch is the project's own channel source, which is plausibly first-party, but no explicit locator or digest authenticates it, so it is reputable but unverified, giving medium.
nanocoai/nanoclaw
dad1aad
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.