CtrlK
BlogDocsLog inGet started
Tessl Logo

llm-application-dev-prompt-optimize

You are an expert prompt engineer specializing in crafting effective prompts for LLMs through advanced techniques including constitutional AI, chain-of-thought reasoning, and model-specific optimizati

46

Quality

47%

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/llm-application-dev-prompt-optimize/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

50%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 body is lean and cleanly sectioned but offers little executable guidance, delegating concrete patterns to a referenced playbook that is not present in the bundle. Validation is mentioned only in the abstract, and the missing reference undermines the otherwise good progressive-disclosure structure.

Suggestions

Provide at least one concrete, copy-paste-ready prompt-optimization example or template inline (e.g. a before/after prompt rewrite) so the skill is actionable without the missing playbook.

Add explicit validation checkpoints — e.g. 'After rewriting, test the prompt against 2-3 representative inputs and confirm the output meets the stated constraints before delivering' — to lift workflow clarity above 3.

Either ship the referenced `resources/implementation-playbook.md` or remove the broken reference; a one-line 'See X' that points to nothing breaks navigation and caps progressive disclosure.

DimensionReasoningScore

Conciseness

The body is lean and well-sectioned without explaining concepts Claude already knows, but it repeats the frontmatter role sentence verbatim and the Context stats ('improve accuracy by 40%... cut costs by 50-80%') are minor fluff that could be trimmed, fitting 'efficient; minor instances of over-explanation' rather than a clean 5.

4 / 5

Actionability

Instructions are abstract ('Clarify goals, constraints, and required inputs', 'Apply relevant best practices and validate outcomes') with no concrete code, commands, or examples; the only concrete pointer is to an external playbook, matching 'minimal concrete guidance; high-level hints but missing the specific steps to execute'.

2 / 5

Workflow Clarity

A rough four-step sequence exists (clarify -> apply best practices & validate -> provide actionable steps & verification -> open playbook) but validation is only mentioned abstractly with no explicit checkpoints, fitting 'steps listed but validation gaps; checkpoints missing or implicit'.

3 / 5

Progressive Disclosure

Sections are well-organized and the playbook reference is one-level-deep and clearly signaled in both Instructions and Resources, but the referenced `resources/implementation-playbook.md` does not exist (no resources/ directory), so the overview's detailed content points to a dead end, fitting the 'some structure but references not effectively delivering' band rather than a 4.

3 / 5

Total

12

/

20

Passed

Description

45%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 the domain and a few named techniques but is role-framed in second person and lacks any explicit 'Use when...' trigger guidance. It is truncated ('optimizati') and does not list concrete deliverable actions, capping its completeness and specificity.

Suggestions

Rewrite the description in third person (e.g. 'Optimizes and crafts effective prompts for LLMs...') instead of 'You are an expert prompt engineer', which both fixes the voice penalty and reads as a capability statement.

Add an explicit trigger clause such as 'Use when the user asks to improve, refine, or optimize prompts for an LLM' to satisfy the 'when' half of completeness.

Replace generic phrases with concrete actions (e.g. 'drafts, restructures, and tests prompts') and include natural synonyms like 'improve' or 'refine' to broaden trigger-term coverage, and fix the truncated 'optimizati'.

DimensionReasoningScore

Specificity

The description names the domain and lists specific techniques ('constitutional AI, chain-of-thought reasoning, and model-specific optimization') but the only concrete action is the generic 'crafting effective prompts'; base specificity is ~3, reduced by 1 because the description uses second person ('You are an expert prompt engineer') per the rubric's voice penalty.

2 / 5

Completeness

It gives a clear 'what' ('crafting effective prompts for LLMs through advanced techniques...') but has no 'Use when...' trigger clause, so completeness is capped at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

It includes a couple of natural terms ('prompt engineer', 'prompt optimization', 'LLMs') but leans heavily on technical jargon ('constitutional AI', 'chain-of-thought') and misses common synonyms like 'improve' or 'refine prompts', matching the 'some relevant keywords but missing common variations' anchor.

3 / 5

Distinctiveness Conflict Risk

Prompt engineering is a recognizable niche, but the role-framed, somewhat generic phrasing ('expert prompt engineer... crafting effective prompts') leaves overlap risk with broader LLM-assistance skills, fitting the 'somewhat specific but could still overlap' anchor.

3 / 5

Total

11

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
rmyndharis/antigravity-skills
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.