CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-orchestration-improve-agent

Systematic improvement of existing agents through performance analysis, prompt engineering, and continuous iteration.

31

Quality

24%

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 ./.agent/skills/agent-orchestration-improve-agent/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

35%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This skill functions primarily as a table of contents pointing to 17 sub-skill files, but the top-level content itself is too abstract and vague to be useful on its own. It lacks any concrete code, commands, examples, or specific techniques—everything actionable is deferred to sub-files that aren't provided. The workflow steps are high-level platitudes rather than executable guidance.

Suggestions

Add at least one concrete, executable example at the top level (e.g., a specific prompt evaluation script, a metric collection command, or a before/after prompt improvement example) so the skill is actionable even without sub-files.

Organize the 17 sub-skill links under clear phase headers (Phase 1: Performance Analysis, Phase 2: Prompt Engineering, Phase 3: Testing & Evaluation, Phase 4: Deployment) instead of a flat numbered list.

Add explicit validation checkpoints to the 4-step workflow, such as 'Verify baseline metrics meet minimum sample size before proceeding' and 'Run regression suite; only deploy if pass rate >= threshold'.

Remove the extended thinking block and 'Use/Do not use' sections to save tokens, and replace with concrete decision criteria or a quick-start example.

DimensionReasoningScore

Conciseness

The extended thinking block and some section headers add unnecessary verbosity. The 'Use this skill when' / 'Do not use this skill when' sections explain things Claude could infer. However, the core instructions are reasonably brief.

2 / 3

Actionability

The skill provides only vague, abstract guidance ('Establish baseline metrics,' 'Identify failure modes,' 'Apply prompt and workflow improvements') with no concrete code, commands, specific examples, or executable steps. Everything actionable is deferred to 17 sub-skill files that are not provided.

1 / 3

Workflow Clarity

There is a 4-step high-level sequence and safety notes about regression testing and rollback. However, there are no validation checkpoints, no feedback loops, and no concrete criteria for when to proceed or roll back. The workflow is too abstract to guide execution reliably.

2 / 3

Progressive Disclosure

The skill references 17 sub-skill files organized by phase, which is a reasonable structure. However, no bundle files are provided, so we cannot verify these references exist. The sub-skills are listed as a flat numbered list without clear grouping headers for phases 2-4, and the 'Phase 1' header awkwardly sits above all 17 links. Navigation could be better organized.

2 / 3

Total

7

/

12

Passed

Description

14%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

This description relies heavily on abstract buzzwords without specifying concrete actions or when the skill should be triggered. It lacks a 'Use when...' clause and fails to differentiate itself from other skills related to prompt engineering, agent development, or optimization. The vague language makes it difficult for Claude to reliably select this skill in the right context.

Suggestions

Add a 'Use when...' clause with specific trigger scenarios, e.g., 'Use when the user wants to improve an existing agent's performance, debug agent behavior, refine prompts, or run evaluations.'

Replace abstract language with concrete actions, e.g., 'Analyzes agent outputs against expected results, rewrites system prompts, adds few-shot examples, designs evaluation harnesses, and iterates on agent configurations.'

Include natural trigger terms users would say, such as 'optimize agent', 'improve prompts', 'agent not working well', 'run evals', 'benchmark agent performance', 'prompt tuning'.

DimensionReasoningScore

Specificity

The description uses vague, abstract language like 'systematic improvement', 'performance analysis', and 'continuous iteration' without listing concrete actions. These are buzzwords rather than specific capabilities.

1 / 3

Completeness

The 'what' is vaguely stated and the 'when' is entirely missing. There is no 'Use when...' clause or equivalent explicit trigger guidance, which per the rubric should cap completeness at 2, but since the 'what' is also weak, this scores a 1.

1 / 3

Trigger Term Quality

Contains some relevant keywords like 'agents', 'prompt engineering', and 'performance analysis' that users might mention, but misses common variations like 'optimize prompts', 'debug agent', 'improve accuracy', 'eval', 'benchmarks', or 'agent tuning'.

2 / 3

Distinctiveness Conflict Risk

The description is very generic and could overlap with many skills related to prompt engineering, code optimization, debugging, or general agent development. 'Systematic improvement' and 'continuous iteration' are too broad to carve out a clear niche.

1 / 3

Total

5

/

12

Passed

Validation

90%

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

Validation10 / 11 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

Total

10

/

11

Passed

Repository
Dokhacgiakhoa/antigravity-ide
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.