CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-orchestration-multi-agent-optimize

Optimize multi-agent systems with coordinated profiling, workload distribution, and cost-aware orchestration. Use when improving agent performance, throughput, or reliability.

51

Quality

56%

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/antigravity-agent-orchestration-multi-agent-optimize/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

42%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 presents a plausible optimization framework but delivers it through buzzword-heavy prose, contentless bullet sections, and non-executable pseudocode. The opening use/do-not-use and Instructions sections are the strongest parts; the rest dilutes them with marketing language and stub code that a competent model cannot run or act on directly.

Suggestions

Cut the 'Context', 'Core Capabilities', and bullet-only sections (Key Strategies, Performance Acceleration, Optimization Spectrum, Observability Framework) — they restate the description in buzzwords without adding actionable content, and replace the role-play framing with direct instructions.

Make code examples executable or remove them: define or drop the undefined classes/functions (DatabasePerformanceAgent, semantic_truncate, PriorityQueue, PerformanceTracker) and replace the 'pass'-bodied CostOptimizer with a working model-selection snippet.

Turn 'Validate improvements with repeatable tests and rollbacks' into an explicit validation checkpoint (e.g., a concrete test command or comparison step with a fix-and-retry loop) in the Instructions sequence, and move the reference workflows and profiling-agent details into files under references/ with clearly signaled links.

DimensionReasoningScore

Conciseness

The body is heavily padded with marketing prose ('Leveraging cutting-edge AI orchestration techniques, this tool provides a comprehensive approach') and contentless bullet lists ('Predictive caching, Pre-warming agent contexts, Intelligent result memoization') that add no actionable information. It stops short of score 1 because it does not explain concepts Claude already knows; it is empty rather than redundant.

2 / 5

Actionability

Every code example is structured pseudocode rather than executable code: 'DatabasePerformanceAgent', 'semantic_truncate', 'PriorityQueue', and 'PerformanceTracker' are undefined, and 'select_optimal_model' is a 'pass' stub. It is above score 2 because the code sketches (ThreadPoolExecutor fan-out, cost model dict) give real shape to the guidance, but nothing is copy-paste ready.

3 / 5

Workflow Clarity

The Instructions section lists a coherent sequence ('Establish baseline metrics... Profile agent workloads... Apply orchestration changes... Validate improvements with repeatable tests and rollbacks'), but validation is a generic step rather than an explicit checkpoint with a fix-and-retry loop. The Safety section's 'regression testing' and 'roll out changes gradually' are named but never operationalized, which is exactly the anchor-3 pattern of sequence present with implicit checkpoints.

3 / 5

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ are absent) and the body contains no references to any external file; all ~240 lines are inline under section headers. Structure exists (8 numbered sections, use/do-not-use sections), but content that clearly belongs in separate files (reference workflows, profiling agent specs, code examples) is inlined, matching anchor 3 rather than 4.

3 / 5

Total

11

/

20

Passed

Description

70%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 solid description with an explicit what and an explicit 'Use when' trigger clause using natural language. Its main weakness is specificity — the capability list reads as orchestration buzzwords rather than concrete operations — though it is clearly distinguishable from neighboring skills.

Suggestions

Replace jargon actions with concrete operations, e.g., 'Profile agent workloads to find bottlenecks, redistribute tasks across agents, and cut token/API costs' instead of 'coordinated profiling, workload distribution, and cost-aware orchestration'.

Broaden the trigger terms with common variations users would say, such as 'agent coordination', 'latency', or 'LLM costs', to lift trigger coverage and distinctiveness.

DimensionReasoningScore

Specificity

'coordinated profiling, workload distribution, and cost-aware orchestration' names the domain and three actions, but they are optimization jargon rather than concrete operations a user can picture. It is not score 4 because, unlike 'Extracts text from PDF files, fills forms, converts pages to images', none of the stated actions is truly specific.

3 / 5

Completeness

Both parts are explicit: the what ('Optimize multi-agent systems with coordinated profiling, workload distribution, and cost-aware orchestration') and the when ('Use when improving agent performance, throughput, or reliability'). It is not score 5 because the what leans buzzwordy and the when-clause covers a narrower set of concrete triggers than the anchor's 'PDF files, PDFs, forms, document extraction, .pdf' style coverage.

4 / 5

Trigger Term Quality

'Use when improving agent performance, throughput, or reliability' uses natural phrases users would say, reinforced by 'multi-agent systems' in the what-clause. It falls short of score 5 because common variations like 'agent coordination', 'latency', and 'cost' are missing from the trigger set.

4 / 5

Distinctiveness Conflict Risk

'multi-agent systems' carves out a clear niche with distinct triggers around agent performance and throughput, so wrong-skill triggering is unlikely. It is not score 5 because it could still overlap with general performance-tuning or cost-optimization skills that mention agents incidentally.

4 / 5

Total

15

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
boisenoise/skills-collections
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.