agent-orchestration-improve-agent

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

1.24x

Quality

51%

Does it follow best practices?

Impact

81%

1.24x

Average score across 3 eval scenarios

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./plugins/AI-Agents-Safe-Coding-Skills/skills/agent-orchestration-improve-agent/SKILL.md

The canonical home for this skill is agent-orchestration-improve-agent in rmyndharis/antigravity-skills

Quality

Content

48%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 skill lays out a thorough, well-sequenced optimization workflow with genuine validation and rollback loops, but it is verbose and relies on pseudocode invocations of non-existent tools rather than executable guidance. Trimming concept explanations and pointing to reference files would materially improve it.

Suggestions

Remove the duplicate intro line and the 'Extended thinking' paragraph, and cut the chain-of-thought / few-shot / Constitutional AI concept explanations that Claude already knows, to recover token budget.

Replace the 'Use: context-manager' / 'Use: prompt-engineer' template blocks with concrete, runnable commands or scripts (or clearly mark them as illustrative invocations with the actual tool/schema to call).

Move the detailed metric catalogs, test-framework specs, and rollout/rollback runbooks into separate reference files (e.g. METRICS.md, TEST-SUITE.md, ROLLOUT.md) and link to them from SKILL.md to improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body restates the description verbatim in the intro, carries a padding 'Extended thinking' meta-paragraph, and explains concepts Claude already knows (chain-of-thought, few-shot examples, Constitutional AI), matching the 'noticeably verbose; several unnecessary explanations or padded sections' anchor.	2 / 5
Actionability	It provides concrete metric catalogs, test categories, and rollback thresholds, but the 'code' blocks are template invocations of fictional tools ('Use: context-manager', 'Use: prompt-engineer') rather than executable commands, fitting the 'some concrete guidance but pseudocode instead of executable code' anchor.	3 / 5
Workflow Clarity	A clearly sequenced four-phase workflow (analysis, prompt engineering, testing, deployment) with validation (A/B testing, statistical-significance gates) and feedback loops (rollback triggers, staged rollout, monitoring) matches the 'clear sequence with most checkpoints present; minor validation gaps' anchor.	4 / 5
Progressive Disclosure	Headers and sub-sections are well organized, but the ~350-line skill is monolithic with no external references and inlines detailed catalogs (metrics, test frameworks) that could live in separate files; since it is well over 50 lines, the simple-skill exception does not apply.	3 / 5
	Total	12 / 20 Passed

Description

53%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 conveys a clear purpose and a reasonably distinct niche, but it is generic in its action vocabulary and entirely lacks explicit trigger guidance. Adding a concrete 'Use when...' clause and sharper natural trigger terms would lift the lower dimensions.

Suggestions

Append an explicit trigger clause, e.g. 'Use when improving an existing agent's performance, reliability, or prompt quality, or when running agent A/B tests and eval suites.'

Replace abstract process terms ('continuous iteration') with concrete user-facing actions ('run A/B eval suites', 'roll back a regressed prompt') to sharpen specificity and trigger terms.

Add natural synonyms users would actually say ('agent tuning', 'prompt debugging', 'eval harness') to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	The description names the domain ('existing agents') and three actions ('performance analysis', 'prompt engineering', 'continuous iteration'), but these are abstract process terms rather than concrete operations, matching the anchor that lists 1-2 concrete actions without being comprehensive.	3 / 5
Completeness	It clearly states what the skill does (systematic agent improvement via three methods) but provides no 'when' guidance; per the rubric, a missing 'Use when...' clause caps completeness at 3.	3 / 5
Trigger Term Quality	Relevant keywords like 'improve', 'agent performance', and 'prompt engineering' appear, but natural variations and synonyms are missing and there is no 'Use when...' trigger phrasing, fitting the 'some relevant keywords but missing common variations' anchor.	3 / 5
Distinctiveness Conflict Risk	'Improvement of existing agents' is a fairly distinct niche with only minor overlap risk against general prompt-engineering or agent-building skills, matching the 'mostly distinct; minor overlap' anchor rather than the broader score-3 anchor.	4 / 5
	Total	13 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: plugins/AI-Agents-Safe-Coding-Skills/skills/agent-orchestration-improve-agent/SKILL.md
Commit: 900c504

Reviewed: 16 minutes 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.