jumpserver-agent-skill — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited jumpserver-agent-skill (Agent Skill) and scored it 92/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 2 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 2 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Use jumpserver-ssh-mcp as the capability layer and this skill as the operating playbook. The human should only need to provide a gateway name, a JumpServer SSH login command, and a target host; the Agent owns configuration, probe, smoke test, and matcher repair.
The Agent must perform the operational setup work itself. Do not ask the human to hand-write profile internals or matcher JSON unless they explicitly want to.
Human-provided inputs should stay small:
Agent-owned outputs:
SSH_ASSIST_PROFILE profile fileBuilt-in matchers are packaged inside jumpserver-ssh-mcp and are available after installation. Agent-authored matchers should be written only to configured custom matcher directories, usually ~/jumpserver-ssh-mcp/matchers/.
Ask for only missing essentials:
gateway_name, such as pro-jumpserver or test-jumpserverssh -i ~/.ssh/pro.pem [email protected] -p2222Optional:
root, __su, or deployprod or testreferences/mcp-client-config.md and configure it.references/profile-authoring.md before creating or changing a profile file.references/onboard-gateway.md before probing a gateway.type: interactive_expect unless maintaining legacy config.matcher: builtin-generic unless a known reference matcher fits better.ssh.matcher_list to verify available matchers and configured gateways.ssh.matcher_probe with the target host and gateway. This must not execute a remote command.ssh.run_command, such as hostname; whoami; id.ssh.matcher_validate, replay transcript snippets with ssh.matcher_test_transcript, then rerun ssh.matcher_probe.ssh.run_command reports a safety confirmation requirement, stop and ask the human before continuing.ssh.file_push / ssh.file_pull; do not use ssh.rsync_upload / ssh.rsync_download with connection_mode: gateway.ssh.file_push / ssh.file_pull to files up to 50MB. For larger artifacts, prefer remote curl / wget, object storage, or direct SSH rsync when available.jumpserver-ssh-mcp://docs/matchers/guidejumpserver-ssh-mcp://docs/matchers/examplejumpserver-ssh-mcp://docs/matchers/schemajumpserver-ssh-mcp://docs/matchers/troubleshootingSSH_ASSIST_PROFILE.ssh.matcher_probe and create a custom matcher.preferred_account in the gateway profile or matcher context.gateway in every command instead of relying on default gateway.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.