agent-topology-optimizer

Agent skill for topology-optimizer - invoke with $agent-topology-optimizer

1.58x

Quality

22%

Does it follow best practices?

Impact

92%

1.58x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/agent-topology-optimizer/SKILL.md

Quality

Content

25%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 a monolithic dump of speculative, non-executable JavaScript that re-implements algorithms Claude already knows, with no progressive disclosure (no bundle files at all), no validation checkpoints for risky topology migrations, and a stray duplicate frontmatter block. The only genuinely actionable material is the short CLI command section. A rewrite as a concise overview pointing to executable reference files would cut the token cost by roughly 90%.

Suggestions

Cut the body to a concise overview (agent role, when to act, decision heuristics for choosing a topology) and move the algorithm implementations, metrics catalogs, and MCP integration patterns into separate reference files under references/, linked one level deep.

Delete the duplicate second YAML frontmatter block, the 'Agent Profile' section that restates the description, and the textbook genetic-algorithm/simulated-annealing boilerplate — describe the algorithm choice criteria instead of re-implementing the algorithms.

Add an explicit validate-before-migrate workflow step (e.g., verify swarm status after `topology-optimize`, with a rollback path) since topology migration is a risky batch operation with estimated downtime.

Dimension	Reasoning	Score
Conciseness	The ~800-line body is severely padded: it includes textbook genetic-algorithm and simulated-annealing implementations Claude already knows, speculative class skeletons, a duplicate YAML frontmatter block, and a redundant 'Performance Focus'/'Specialization' restatement of the description. Almost none of this earns its place in the context window.	1 / 5
Actionability	The 'Operational Commands' section provides concrete `npx claude-flow` CLI invocations, but the ~700 lines of JavaScript are pseudocode: every class depends on undefined collaborators (TopologyAnalyzer, GeneticAlgorithm, mcp.*) with no setup, imports, or runnable path. This matches anchor 3 ('pseudocode instead of executable code') better than anchor 2 (there is real concrete guidance) or anchor 4 (nothing is copy-paste executable).	3 / 5
Workflow Clarity	A rough analyze→generate→evaluate→select→migrate sequence exists inside optimizeTopology, but there are no validation checkpoints or error-recovery loops anywhere, despite topology migration being a risky batch operation with 'estimatedDowntime'. This fits anchor 2 (rough sequence, validation absent); the absence of any validate-then-proceed gate also caps this dimension at 3 by the rubric's feedback-loop rule.	2 / 5
Progressive Disclosure	There are no references/, scripts/, or assets/ directories — everything (full algorithm implementations, metrics catalogs, MCP integration patterns) is inlined into a single monolithic SKILL.md with zero external references. Section headers exist (above anchor 1), but content that clearly belongs in separate files is entirely inline, matching anchor 2.	2 / 5
	Total	8 / 20 Passed

Description

20%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 frontmatter description is a placeholder: it tells the agent how to invoke the skill but not what it does or when to use it, which would leave the skill undiscoverable for real user requests. Notably, the more informative description ('Dynamic swarm topology reconfiguration and communication pattern optimization') is stranded in a second, duplicate YAML frontmatter block inside the body rather than in the actual frontmatter field. Even that second line lacks a 'Use when...' trigger clause.

Suggestions

Replace the frontmatter description with a capability statement plus trigger guidance, e.g., 'Optimizes swarm topology by reconfiguring network structures, placing agents, and optimizing communication patterns. Use when the user mentions swarm topology, network latency between agents, agent placement, or communication overhead.'

Remove the duplicate second YAML frontmatter block from the body and keep a single, valid frontmatter with the real description.

Include natural trigger terms and synonyms users would actually say ('topology', 'latency', 'agent placement', 'mesh', 'hierarchical') so the skill is selected over sibling optimization agents.

Dimension	Reasoning	Score
Specificity	The description 'Agent skill for topology-optimizer - invoke with $agent-topology-optimizer' names the domain but states no capability actions at all, only how to invoke the skill. It is above anchor 1 because the domain is explicitly named, but below anchor 3 because zero concrete actions (e.g., 'reconfigures swarm topology') are described.	2 / 5
Completeness	Neither 'what' (no statement of what the skill does) nor 'when' (no trigger guidance of any kind) is present — the description is purely meta-invocation text. This matches anchor 1; anchor 2 would require at least a vague 'what' or a 'when', neither of which exists, and the missing 'Use when' clause caps completeness at 3 regardless.	1 / 5
Trigger Term Quality	The only recognizable keyword is 'topology-optimizer'; the rest is invocation syntax ('$agent-topology-optimizer') that no user would naturally say. Not score 1 because one domain keyword exists, but it lacks the natural phrases and synonyms (e.g., 'swarm topology', 'network optimization') needed for score 3 or higher.	2 / 5
Distinctiveness Conflict Risk	The niche name 'topology-optimizer' provides some specificity, but the generic 'Agent skill for X - invoke with $X' template gives no distinguishing trigger surface. It is more specific than anchor 2's broad overlap risk, but cannot reach anchor 4 without concrete triggers that separate it from sibling optimization agents.	3 / 5
	Total	8 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (813 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: ruvnet/ruflo
Path: .agents/skills/agent-topology-optimizer/SKILL.md
Commit: b14c79e

Reviewed: 3 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.