CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ax-python-agent-rlm

Use when writing Python code with `axllm` for RLM executor loops, AxCodeRuntime sessions, runtime envelopes, process runtimes, and optional runtime profiles.

Quality

83%

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

SKILL.md
Quality
Evals
Security

Quality

Content

82%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 tight, well-structured reference skill that respects the token budget and provides a concrete core code pattern with useful guardrail checkpoints. Its main weaknesses are an example that is not fully self-contained (`llm` undefined), no code for the runtime/profile use cases it headlines, and referenced files that are not present in the bundle.

DimensionReasoningScore

Conciseness

The body is lean and bullet-driven ("Real network support: yes.", "Scripted no-key transport support: yes.") with a one-line intro and no explanation of concepts Claude already knows. Every token earns its place, matching anchor 5 rather than 4, which requires identifiable over-explanation to trim.

5 / 5

Actionability

The Core Pattern supplies real executable code (`helper = agent("question:string -> answer:string")`, `out = helper.forward(llm, {...})`), but `llm` is never constructed and headline use cases like `ProcessCodeRuntime` and runtime profiles get names only, no code. Mostly executable with minor gaps — anchor 4, not 5 (not copy-paste self-contained) and not 3 (it is executable code, not pseudocode).

4 / 5

Workflow Clarity

There is no sequenced multi-step workflow, but the Guardrails function as explicit decision checkpoints ("Start from package examples for exact native syntax…", "Use `no-key` examples for deterministic local checks…") and the skill involves no destructive or batch operations. Anchor 4 rather than 5 because the path across the several listed scenarios is not unified, and not 3 because checkpoints are explicit rather than implicit.

4 / 5

Progressive Disclosure

A well-organized sub-50-line overview with clearly labeled one-level-deep pointers ("Package API docs: `API.md` and `axir-api.json`", "Runnable examples: `examples/`"). However, none of the referenced files exist in the skill bundle, so navigation cannot be confirmed — anchor 4 rather than 5, and not 3 because the structure is clean and references are clearly signaled rather than buried.

4 / 5

Total

17

/

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 focused, jargon-accurate description with an explicit 'Use when' trigger and a highly distinctive niche. Its main limits are a single action verb stretched over a list of objects, missing natural trigger variations, and a when-clause that is circular with the what.

Suggestions

Add natural trigger variations users might actually say (e.g. 'AxCode', 'agent runtime', 'Pyodide', 'QuickJS') so the description matches more phrasings.

Make the 'when' clause add concrete triggers beyond restating the 'what' (e.g. 'Use when the user mentions axllm, AxCodeRuntime, or runtime profiles').

DimensionReasoningScore

Specificity

Lists five concrete work areas ("RLM executor loops, AxCodeRuntime sessions, runtime envelopes, process runtimes, and optional runtime profiles") under a single action verb ("writing Python code with `axllm`"). This matches anchor 4 — several specific items with minor gaps — rather than 5, because the items are objects of one action rather than multiple distinct concrete actions.

4 / 5

Completeness

Both parts are present: the what ("writing Python code with `axllm` for RLM executor loops, …") and an explicit "Use when…" clause. Not 5 because the when-clause merely restates the what without adding concrete trigger phrases; not 3 because the when is explicit, not merely implied.

4 / 5

Trigger Term Quality

"writing Python code", "axllm", "AxCodeRuntime sessions", and "runtime profiles" are terms a user of this package would naturally say. Good coverage but missing natural variations (e.g. "AxCode", "agent runtime", "Pyodide", "QuickJS"), so anchor 4 rather than 5.

4 / 5

Distinctiveness Conflict Risk

"axllm", "RLM executor loops", and "AxCodeRuntime" define a clear niche with distinct triggers and minimal conflict risk with any generic Python skill. Matches anchor 5; the package qualifier prevents even the minor overlap implied by anchor 4.

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

Path

packages/python/skills/ax-python-agent-rlm/SKILL.md

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.