CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

working-across-computers

Guides work across Cloud, local, and other available computers, including teleporting conversations, running remote subagents, and transferring files. Load when a request needs the user's laptop or another machine's files, apps, or tools; when connecting a computer; or when moving or delegating work between local and Cloud in either direction, including running Cloud subagents from local.

Quality

95%

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

SKILL.md
Quality
Evals
Security

Quality

Content

100%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 skill body is an exemplary instruction-style skill: lean, concrete commands for every workflow, explicit validation and error-recovery steps for the risky teleport/transfer operations, and a clean split of connection setup into a single one-level reference. No padding, no known-concept explanations, and no organizational weaknesses were found.

DimensionReasoningScore

Conciseness

The body is dense with non-obvious, domain-specific guidance ('never reconstruct it from the filename', 'Download sources must be under /root/downloads', 'Target flags override the executing session without changing its identity') and explains nothing Claude already knows. Every token earns its place, matching the anchor-5 example of lean, competence-assuming content.

5 / 5

Actionability

Fully executable commands cover every flow: `letta sandbox upload/download` shown in all three target-flag variants, `letta teleport cloud`, `letta teleport list`, `letta teleport <computer>`, and `desktop-link`, plus an intent-to-mechanism table. This is copy-paste-ready guidance covering the common cases, matching anchor 5.

5 / 5

Workflow Clarity

Multi-step flows are clearly sequenced with explicit validation checkpoints and error-recovery loops: transfer files before teleporting, 'verify completion before moving', 'If teleport fails, stay on the source and resolve the error before retrying', 'verify received contents or checksums', and running the teleport handoff alone as the final tool call. This matches anchor 5's explicit validation with feedback loops; no gaps found.

5 / 5

Progressive Disclosure

The body stays a concise overview of routing, teleport, transfer, and desktop-link workflows, while first-time computer-connection details are appropriately split into a single real, well-signaled, one-level-deep reference (`references/connect-a-computer-for-remote-access.md`, linked twice with clear trigger conditions). Structure and navigation match anchor 5.

5 / 5

Total

20

/

20

Passed

Description

87%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.

A strong description that clearly states both capabilities and explicit load triggers in third-person voice. The only notable gap is that the sandbox desktop-view sharing capability from the body is not mentioned, leaving slightly incomplete coverage of what the skill does.

Suggestions

Add the sandbox desktop-view sharing capability (e.g., 'sharing a live view of a Cloud sandbox desktop') to the capability list to make coverage comprehensive.

Include a couple more natural trigger synonyms such as 'remote machine' or 'another computer' to broaden natural keyword coverage.

DimensionReasoningScore

Specificity

Enumerates several concrete actions — 'teleporting conversations, running remote subagents, and transferring files' plus 'connecting a computer' — but omits the sandbox desktop-view sharing capability covered in the body, leaving minor gaps in coverage. It exceeds the 3 anchor (which allows only 1-2 actions) but falls short of anchor 5's comprehensive coverage.

4 / 5

Completeness

The first sentence explicitly states what the skill does and the second gives concrete trigger conditions ('Load when a request needs the user's laptop...; when connecting a computer; or when moving or delegating work... in either direction'). Both what and when are explicit with concrete trigger phrases, matching anchor 5 exactly.

5 / 5

Trigger Term Quality

Natural user phrases are present: 'the user's laptop or another machine's files, apps, or tools', 'connecting a computer', 'moving or delegating work between local and Cloud'. A few natural terms are missing (e.g., 'remote machine', 'desktop', 'workstation'), matching anchor 4's 'good keyword coverage; a few natural terms missing' rather than anchor 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

It carves a clear niche — cross-computer orchestration via Letta-specific mechanisms (teleport, Cloud sandbox, remote subagents) — with trigger conditions unlikely to fire for unrelated skills. Anchor 5 ('clear niche with distinct triggers; minimal conflict risk') is the best fit; no neighboring anchor matches better.

5 / 5

Total

18

/

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

letta-ai/letta-code

Path

src/skills/builtin/working-across-computers/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.