CtrlK
BlogDocsLog inGet started
Tessl Logo

ax-python-gepa

Use when writing Python code with `axllm` for GEPA, Pareto tradeoffs, reflection clients, metric budgets, optimizer state, and artifacts.

58

Quality

73%

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 ./website/static/python/.well-known/agent-skills/ax-python-gepa/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

57%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.

A concise, well-organized overview with useful package facts and sensible guardrails, but its guidance is more descriptive than executable: the code sample is a placeholder call shape, no workflow is sequenced with validation checkpoints, and the cited docs/examples are not clearly navigable references.

Suggestions

Replace or augment the Core Pattern with one complete, runnable example (real request/evaluator construction or a pointer to a specific file in `examples/`) so the primary code sample is copy-paste executable.

Add a short numbered workflow for the main use cases (e.g. seed with AxBootstrapFewShot → run AxGEPA.optimize → inspect Pareto front/state) with an explicit verification step for results.

Turn the 'Package Facts' entries into clearly signaled links ('API reference: see [API.md](API.md)') and, if the cited files belong to the generated package rather than the skill bundle, state where to find them.

DimensionReasoningScore

Conciseness

The ~40-line body is lean and assumes competence: no explanation of what GEPA or optimization is, and the facts lines ('Real network support: yes', 'Runtime profiles: ...') each carry information. Minor redundancy remains — the intro paragraph restates the description, and 'Relevant API Surface' overlaps 'Package Facts' — so it sits at anchor 4 rather than 5.

4 / 5

Actionability

The Core Pattern block ('engine = AxGEPA(reflection_client)', 'engine.optimize(request, evaluator)') uses undefined placeholders — a call shape, not executable code — though it points to `examples/` and guardrails direct 'Start from package examples for exact native syntax'. Concrete but incomplete with missing key details matches anchor 3 better than anchor 4's 'mostly executable' guidance.

3 / 5

Workflow Clarity

'When To Use' lists scenarios rather than steps; the only sequencing signal is 'Use BootstrapFewShot before GEPA when demonstrations should seed optimization'. There is no ordered workflow and no validation checkpoints, though the operations are not destructive. Above anchor 2 (guidance is coherent and guardrails give decision rules), below anchor 4 (no clear sequence with checkpoints).

3 / 5

Progressive Disclosure

The body cites `API.md`, `axir-api.json`, `axir-capabilities.json`, and `examples/`, but no bundle files exist in the skill directory, and the citations appear as bare facts under 'Package Facts' rather than clearly signaled navigation links. Sections are well organized, but the references are neither verifiable in the bundle nor clearly signposted — anchor 3.

3 / 5

Total

13

/

20

Passed

Description

78%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 strong, niche-specific description with an explicit 'Use when' trigger clause and good domain keyword coverage. Its main weakness is that the 'what' is only implied by the enumerated topics rather than stated as concrete actions, and it offers no synonym variations.

Suggestions

Add a leading 'what' clause stating the concrete actions, e.g. 'Run GEPA optimizers and Pareto-front analysis... Use when writing Python code with `axllm`...'.

Include natural synonym variations such as 'optimizer', 'candidate evaluation', and 'ax package' to broaden trigger coverage.

DimensionReasoningScore

Specificity

Names the domain (Python, `axllm`) and enumerates several concrete capability areas — 'GEPA, Pareto tradeoffs, reflection clients, metric budgets, optimizer state, and artifacts' — with minor coverage gaps. These are topics rather than explicit actions like 'run' or 'track', so it does not reach anchor 5's comprehensive concrete-action coverage.

4 / 5

Completeness

The explicit 'Use when writing Python code with `axllm` for...' clause answers 'when', and the enumerated topics weakly imply the 'what', but there is no separate, explicit statement of what the skill does. The 'when' is explicit (above anchor 3, where 'when' is missing or weakly implied), but the merged single-clause form falls short of anchor 5's distinct what-and-when with concrete trigger phrases.

4 / 5

Trigger Term Quality

Good keyword coverage of niche terms users of this package would naturally say ('Python', 'axllm', 'GEPA', 'Pareto tradeoffs', 'metric budgets'), but it lacks synonyms and common variations (e.g. 'optimizer', 'ax'). Better than anchor 3's 'some relevant keywords'; short of anchor 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

The highly niche vocabulary ('axllm', 'GEPA', 'Pareto tradeoffs', 'reflection clients', 'optimizer state') forms a clear niche with distinct triggers and virtually no overlap with generic skills. Matches anchor 5; nothing generic that would fire the wrong skill.

5 / 5

Total

17

/

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
ax-llm/ax
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.