Tessl • The management layer for AI agent skills

Blog Docs Log inGet started

SkillPublic

orbit

New remote

Quality

30%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./crates/orbit-cli/tests/fixtures/skills/new-remote/SKILL.md

Quality

Content

30%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 body is an unbuilt template shell: a bare heading plus two `orbit:include` directives whose targets (local:shared, local:new-only) are absent from the bundle, leaving zero instructional content. While maximally lean (hence the high conciseness score), it provides no executable guidance, no workflow, and no working references. This mirrors the rubric's bad_overall_example of a reference chain that resolves to nothing.

Suggestions

Resolve or remove the dangling "orbit:include local:shared" / "local:new-only" placeholders — either build the template so the includes expand into real content, or inline the actual guidance in the body.

Add concrete, executable Orbit CLI commands and a short worked example so actionability reflects what the skill actually does (per the frontmatter, it requires the Orbit CLI — show its usage).

If multi-step operations are involved (create/sync/teardown of remotes), write a numbered sequence with explicit validation checkpoints, and split any bulk material into real reference files listed in the bundle.

Dimension	Reasoning	Score
Conciseness	The body is three lines — a heading ("# Remote") and two HTML-comment include directives — with zero padding, zero explanation of concepts Claude already knows, and no over-explanation to trim, matching anchor 5 ("Lean and efficient; every token earns its place"). Note this leanness reflects emptiness rather than discipline; the resulting absence of any guidance is penalized under actionability, workflow_clarity, and progressive_disclosure, not here.	5 / 5
Actionability	The body contains no concrete code, commands, or instructions whatsoever — only a title and two templating comments ("orbit:include local:shared", "orbit:include local:new-only") that resolve to no files in the bundle. This matches anchor 1 ("Entirely vague or abstract; no concrete code or commands; only describes rather than instructs"); anchor 2 requires at least high-level hints, which are absent.	1 / 5
Workflow Clarity	No steps or sequence exist in any form — the body has no process content at all, matching anchor 1 ("Steps missing or incoherent; no sequence"). Anchor 2 requires a rough sequence to be present, which it is not; there is nothing to validate or order.	1 / 5
Progressive Disclosure	Scored against the actual bundle structure per the judging guidelines: the "orbit:include local:shared" and "orbit:include local:new-only" references point to files that do not exist anywhere in the bundle (no references/, scripts/, assets/, or shared/ directories), so there is no content to navigate to and no usable structure — matching anchor 1 ("impossible to navigate"). Anchor 2 requires either inlined content or buried-but-real references; the references here are broken, which is worse.	1 / 5
	Total	8 / 20 Passed

Description

0%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 "New remote" is a placeholder, not a description: it states no capabilities, contains no trigger terms, answers neither what nor when, and would conflict with many unrelated skills. It matches the rubric's bad_overall_examples category ("Helps with documents", "Does stuff with files") — indeed it is weaker than those. Voice guidance (third person) is not violated but there is effectively no voice at all.

Suggestions

Replace "New remote" with a concrete capability statement naming the actual operations, e.g. "Creates, inspects, and tears down Orbit remote environments via the Orbit CLI" — this lifts specificity from anchor 1 to anchor 4+.

Add an explicit "Use when..." trigger clause with natural user phrasing, e.g. "Use when the user mentions Orbit remotes, syncing to a remote environment, or setting up a new remote" — this is required for completeness to reach 4-5 and supplies trigger terms.

Include distinguishing vocabulary ("Orbit CLI", "Orbit remote") so the description cannot collide with git-remote or generic remote-server skills, addressing distinctiveness/conflict risk.

Dimension	Reasoning	Score
Specificity	The entire description is "New remote" — no concrete actions, no domain beyond the word "remote", pure abstract placeholder language. It exactly matches the score-1 anchor ("Entirely vague; no concrete actions"), and there is nothing present to lift it to anchor 2, which at minimum requires naming the domain with minimal actions.	1 / 5
Completeness	The description answers neither "what does this do" (no capability is stated) nor "when should Claude use it" (no trigger clause, and the judging guidelines explicitly cap a missing 'Use when...' at 3 even when a clear 'what' exists). Both are extremely vague, matching anchor 1; anchor 2 requires at least a vague 'what' or a 'when', neither of which is present.	1 / 5
Trigger Term Quality	"New remote" contains no natural keywords a user would say to invoke this skill — no verbs, no file formats, no synonyms, not even generic ones like "files" or "documents". This matches anchor 1 ("only technical jargon or entirely generic language"); anchor 2 requires at least one or two generic keywords, which are absent.	1 / 5
Distinctiveness Conflict Risk	"New remote" is entirely generic and would collide with virtually any skill involving remotes — git remotes, remote servers, remote sessions, remote work. Anchor 1 ("would conflict with virtually any skill") is the closest fit; even anchor 2's example ("Helps with document files") names a more specific domain than this does.	1 / 5
	Total	4 / 20 Passed

Validation

81%

Warnings & errors only

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

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
description_field	'description' is very short (10 chars), consider making it more detailed	Warning
metadata_version	'metadata.version' is missing	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: gitlabhq/orbit-knowledge-graph
Path: crates/orbit-cli/tests/fixtures/skills/new-remote/SKILL.md
Revision: ed321d9

Reviewed: 27 days ago

Discovery Implementation Validation

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.

orbit

Table of Contents

Is this your skill?