mode-classification

Use before routing a /meta-agent request to choose single-agent-creator, team-builder, or agentlas-packager from the user's wording and available files.

Quality

75%

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 ./.agents/skills/mode-classification/SKILL.md

Quality

Content

71%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.

A well-structured, actionable decision procedure that stays concise and assumes Claude's competence. Its weakest point is progressive disclosure: several referenced files are dangling because no bundle files are present, and worked examples would strengthen actionability.

Suggestions

Provide the referenced bundle files (docs/mode-classifier.md, modes/ontology-backed-agent.md, .agentlas/contract-injection-map.json) or inline the essential detail they would contain so the procedure is self-contained.

Add 2-3 worked examples mapping sample user requests to a chosen mode, overlay, and loop_policy to lift actionability and workflow clarity.

Define the "clarify question loop" inline or point to a real file that defines it, so the feedback checkpoint is executable rather than assumed.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence, explaining only agentlas-specific concepts (ownership boundaries, produces/consumes, orchestrator/HQ) rather than general knowledge; minor tightening is possible in Step 1 and the MULTI/SINGLE hints of Step 6 which overlap earlier steps.	4 / 5
Actionability	As an instruction-only skill it gives concrete if/then decision rules (count boundaries, check synthesis need, shape guard) that are directly executable; it falls short of 5 because it lacks worked request-to-mode examples that would cover common cases.	4 / 5
Workflow Clarity	Steps 0-3 plus overlay/loop-policy steps form a clear numbered sequence with a clarify-loop checkpoint for ambiguous cases; it is not a 5 because the clarify loop and loop-policy risk tiers are referenced but not defined inline.	4 / 5
Progressive Disclosure	Sections are well organized and the Reference section signals one-level-deep pointers, but the referenced files (docs/mode-classifier.md, modes/ontology-backed-agent.md, .agentlas/contract-injection-map.json) do not exist in the bundle, so navigation does not resolve.	3 / 5
	Total	15 / 20 Passed

Description

80%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 concise, third-person description that clearly states both what it does and when to use it, with concrete named targets and a distinct trigger. Its main weakness is jargon-heavy trigger terms that miss natural user phrasings.

Suggestions

Add a few natural-language trigger synonyms users might actually say (e.g., "Use when the user wants to build, repair, or package an agent or agent team") alongside the technical "/meta-agent request" phrasing.

Consider naming the underlying action in plainer terms (e.g., "classify") so the capability reads concretely without relying on the mode names alone.

Dimension	Reasoning	Score
Specificity	The description names the concrete action ("choose single-agent-creator, team-builder, or agentlas-packager") and enumerates three specific output targets, giving concrete coverage of the routing decision with only minor gaps.	4 / 5
Completeness	It explicitly answers both what ("choose single-agent-creator, team-builder, or agentlas-packager from the user's wording and available files") and when ("Use before routing a /meta-agent request") with a concrete trigger clause.	5 / 5
Trigger Term Quality	It includes the relevant trigger phrase "before routing a /meta-agent request" and the three mode names, but these are largely technical jargon and lack the natural synonyms/variations a user might actually say.	3 / 5
Distinctiveness Conflict Risk	The "/meta-agent request" trigger and the three named agentlas modes carve a clear niche with highly distinctive triggers and minimal overlap with other skills.	5 / 5
	Total	17 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: agentlas-ai/Agentlas-OS
Path: .agents/skills/mode-classification/SKILL.md
Commit: a7988bb

Reviewed: 4 days ago

Table of Contents

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.