CtrlK
BlogDocsLog inGet started
Tessl Logo

ax-python-agent-optimize

Use when writing Python code with `axllm` for agent optimization, verified agent-playbook evolution, evaluators, judges, optimizer artifacts, BootstrapFewShot, and GEPA.

60

Quality

75%

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-agent-optimize/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

61%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-sectioned overview with a real API call shape and sensible guardrails, but it stops short of being self-sufficient: the code example is a sketch with undefined variables, the referenced examples/API files are absent from the bundle, and no stepwise workflow with validation checkpoints operationalizes the verification gate and budgets it mentions.

Suggestions

Replace the placeholder Core Pattern with one complete runnable example — construct a real evaluator callback and request (or inline a minimal example from the package's examples/) so the code is copy-paste executable rather than a call sketch with undefined variables.

Add a short numbered workflow for an optimization run (pick a package example → define the evaluator → run the optimizer with an explicit budget/row limit → keep only playbook proposals that pass the verification gate), so the verification gate and budget guardrails become operational checkpoints.

Since the referenced 'API.md', 'axir-api.json', and 'examples/' files are not present in this skill bundle, either bundle the key excerpts (e.g., the AxGEPA/AxBootstrapFewShot signatures) or give resolvable paths to them so the deferral actually lands somewhere.

DimensionReasoningScore

Conciseness

The body is lean — terse fact lists ('Real network support: yes.'), guardrails as one-line directives, and no explanation of concepts Claude already knows. Minor trimmable redundancy remains (the intro paragraph re-states 'generated Ax package `axllm`' and 'Language: Python.' duplicates it again), which keeps it below the leanest anchor.

4 / 5

Actionability

The Core Pattern uses undefined placeholders ('AxGEPA(reflection_client)', 'engine.optimize(request, evaluator)') with no evaluator or request construction shown, and the 'Runnable examples: `examples/`' and API docs it defers to are not present in the bundle, leaving the concrete-execution gaps larger than 'minor'.

3 / 5

Workflow Clarity

No sequenced workflow for an optimization run exists; the verification gate and budgets appear only as guardrail phrases without steps or checkpoints. Optimization runs are batch operations over datasets, and the rubric caps workflow clarity at 3 when the validation loop is not operationalized.

3 / 5

Progressive Disclosure

Sections are well organized and appropriately short for a sub-50-line skill, but pointers such as '`API.md` and `axir-api.json`' and 'examples/' are bare filenames whose target files are not in the bundle, leaving minor navigation and organization gaps.

4 / 5

Total

14

/

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 and concrete package terminology. Its main weaknesses are topic-noun phrasing instead of explicit capability actions and limited synonym coverage around the optimization domain.

DimensionReasoningScore

Specificity

Names the domain ('writing Python code with `axllm`') and several concrete items ('evaluators, judges, optimizer artifacts, BootstrapFewShot, and GEPA'), but these are noun-topics rather than the multiple explicit actions of the top anchor, leaving minor coverage gaps.

4 / 5

Completeness

The 'when' is explicit ('Use when writing Python code with `axllm`...') and the 'what' is conveyed through the concrete topic list, but the what is implied rather than stated as a skill action, so it is not the fully explicit what+when of the top anchor.

4 / 5

Trigger Term Quality

Good coverage of natural niche terms a user of this package would say ('agent optimization', 'evaluators', 'judges', 'BootstrapFewShot', 'GEPA'), though common variations like 'prompt optimization', 'few-shot', or 'optimize an agent' are missing.

4 / 5

Distinctiveness Conflict Risk

Terms like 'axllm', 'BootstrapFewShot', and 'GEPA' define a clear niche with distinct triggers; no plausible confusion with generic Python or agent skills, so conflict risk is minimal.

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.