CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

remote-mac

Remote Macs: MacBooks, Mac Studios, hosted claw Macs, Tailscale, SSH, and OpenClaw.

Quality

70%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/remote-mac/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is highly actionable operational knowledge — exact commands, hostnames, and safety gates — with clear discovery and fallback workflows. Its weaknesses are monolithic inlining of volatile per-host state (which also makes sections go stale), some duplication across sections, and a few prose-only multi-step flows without explicit checkpoints.

Suggestions

Split volatile per-host status (clawmac outage tickets, miniclaw regression, foundationclaw pending reset) into a references/ file (e.g., HOSTS.md) so SKILL.md holds only stable routing rules and the mutable state is one clearly-signaled reference deep.

Deduplicate the Molty-retirement note (appears in both Peter's Topology and OpenClaw Checks) and consolidate the clawmac outage details currently spread across Topology, Discovery, and clawmac GUI Access.

Convert the Codex automation migration and gog OAuth flows into numbered steps with an explicit verification command at the end (e.g., 'gog auth list --check --json --no-input' already exists — surface it as a numbered checkpoint).

DimensionReasoningScore

Conciseness

The body is dense with fleet-specific facts Claude cannot know (hostnames, tails, ticket numbers, approval gates) and explains no general concepts, but it is not fully lean: Molty's retirement is stated twice, the clawmac outage is repeated across three sections, and volatile time-stamped state ('Live 2026-08-01 state regressed', ticket #11386/#11481) sits inline rather than in a status/deprecated section. These are minor instances that could be trimmed, matching anchor 4 rather than 5.

4 / 5

Actionability

The guidance is fully executable: exact hostnames and users, 'ssh -o RequestTTY=no -o RemoteCommand=none HOST \"COMMAND\"', 'dns-sd -B _ssh._tcp local', 'lsof -nP -iTCP:18789 -sTCP:LISTEN', 'openclaw gateway status --deep --require-rpc --json', and 'peekaboo list windows --app \"Jump Desktop\" --json' cover the common discovery, SSH, health-check, and GUI-access cases copy-paste ready. Prose-only passages (approval policy) are inherently policy, not missing commands.

5 / 5

Workflow Clarity

Discovery is an 8-step numbered sequence with explicit checkpoints (verify ComputerName/hardware UUID/node ID before merging peers), fallbacks (Tailscale -> dns-sd/arp -> mDNS), and hard stop conditions; destructive operations (reboots, gateway restarts, reimaging) are explicitly gated behind per-task approval rather than left unvalidated. Not 5: the Codex automation migration and gog OAuth flows are prose descriptions without numbered validation checkpoints.

4 / 5

Progressive Disclosure

The body is well-sectioned (Topology, Discovery, SSH Rules, OpenClaw Checks, Safety), but at roughly 127 lines with no bundle files, volatile per-host status and topology detail is inlined monolithically rather than split into a references file. References to 'computers.yaml' and 'docs/fleet-setup.md' point to a private manager repo outside the skill, not navigable bundle files, so content that should be separate is inline — matching anchor 3.

3 / 5

Total

16

/

20

Passed

Description

47%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is a domain keyword list rather than a capability statement: it names the machines and tooling but never says what the skill does or when to invoke it. It is distinctive within its niche but fails the what/when completeness test that trigger descriptions require.

Suggestions

Add an explicit 'Use when...' clause, e.g., 'Use when the user mentions MacBook, Mac Studio, clawmac, or asks to run, check, or configure something on one of Peter's remote Macs.'

State concrete actions the skill performs, such as 'Run commands over non-interactive SSH, discover hosts via Tailscale and LAN fallback, check OpenClaw gateway health, and configure Codex cron automations on remote Macs.'

Include natural trigger variations users are likely to say, such as 'Mac mini', 'remote into', or 'run on my Mac', to round out keyword coverage.

DimensionReasoningScore

Specificity

The description 'Remote Macs: MacBooks, Mac Studios, hosted claw Macs, Tailscale, SSH, and OpenClaw' names a concrete domain but contains no action verbs whatsoever — it never states what the skill does. It avoids score 1 because the listed entities are specific rather than abstract, but it cannot reach 3 without naming any concrete action.

2 / 5

Completeness

There is only a vague implicit 'what' (dealing with remote Macs) and no 'Use when...' clause or equivalent trigger guidance, matching the anchor 'has a vague what and no when'. The missing 'Use when' clause caps this dimension at 3 regardless.

2 / 5

Trigger Term Quality

Natural keywords users would actually say are present ('MacBooks', 'Mac Studios', 'Tailscale', 'SSH', 'OpenClaw'), giving good coverage. Not 5: common variations and phrases are missing (e.g., 'Mac mini', 'run on my Mac', 'remote into my Mac').

4 / 5

Distinctiveness Conflict Risk

'Remote Macs', 'hosted claw Macs', and 'OpenClaw' establish a clear niche, but the generic terms 'SSH' and 'Tailscale' create minor overlap risk with general remote-access or networking skills, preventing a 5.

4 / 5

Total

12

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

steipete/agent-scripts

Path

skills/remote-mac/SKILL.md

Reviewed

Table of Contents

Is this your skill?

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.