Postwall
ReviewAudited by ClawScan on May 1, 2026.
Overview
Postwall is a coherent email-approval gateway, but it uses a PostWall API key and an npm-installed CLI to let an agent read approved emails and create or update drafts, so users should monitor that access.
Install this only if you intend to give the agent PostWall-mediated email access. Use a revocable API key, verify the npm package and service source, treat email bodies as untrusted content, and review drafts carefully in the PostWall dashboard before approving any send.
Findings (4)
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.
Anyone enabling the skill is delegating PostWall-mediated email access to the agent, including reading approved messages and creating or updating outgoing drafts.
The skill requires a PostWall API key so the agent can perform email-gateway operations on the user's behalf.
"env":["POSTWALL_API_KEY"] ... First, authenticate with your API key
Use a dedicated, revocable PostWall API key, confirm the dashboard approval settings, and rotate or revoke the key if the agent should no longer have access.
An agent could mark approved emails as read or prepare drafts; a human still needs to approve sending, but read-status changes may affect what appears in later inbox checks.
The documented commands can change email state and create or update outgoing drafts, although actual sending is described as human-approval gated.
**This marks the email as read** ... postwall mark-read <id1> ... The email is NOT sent until approved in the dashboard.
Review agent workflows that call read or mark-read, and approve outgoing drafts only after checking recipients, subject, body, and approval URL.
A malicious or misleading email could try to steer the agent if the agent treats email text as instructions instead of data.
The skill brings email content into the agent's working context; email bodies can contain instructions or links from external senders.
postwall read <email-id> # Shows email content
Instruct the agent to treat email contents as untrusted data and to ask before acting on requests, links, credentials, or instructions found inside emails.
The installed CLI will handle the PostWall API key and email operations, so package provenance matters.
Installation depends on an external npm package that provides the executable used by the skill.
[0] node | package: postwall | creates binaries: postwall
Install from the expected npm package, verify the linked homepage or repository, and keep the package updated from trusted sources.