CtrlK
BlogDocsLog inGet started
Tessl Logo

ax-rust-flow

Use when writing Rust code with `axllm` for flows, nodes, program graphs, nested programs, dynamic options, caching, and optimizer components.

60

Quality

76%

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 ./packages/rust/skills/ax-rust-flow/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%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 flow-patterns core of this skill is excellent — lean, concrete, executable Rust examples that build from a core pattern to advanced composition. The score is dragged down by the 'Astra Session Work' section, a dense ~30-line inline spec dump covering five languages and MCP/session machinery that is largely out of scope for a Rust flow skill and should live in its own reference file.

Suggestions

Move the 'Astra Session Work' section into a separate reference file (e.g. sessions.md) and keep a two- to three-line pointer in SKILL.md; this section is the main driver of both the conciseness and progressive-disclosure deductions.

Trim the Astra content to Rust-only facts when it is referenced, since details like 'add_child_agent in Python/C++, AddChildAgent in Go, addChildAgent in Java' are irrelevant to a skill whose stated purpose is writing Rust code with `axllm`.

Add a brief validation step for produced flows (e.g. run the no-key examples or a scripted transport check before handing code over), which would close the workflow-clarity gap between pattern reference and verified output.

DimensionReasoningScore

Conciseness

The flow-patterns half ("Core Pattern" through "Cache a flow") is lean and every token earns its place, but "Astra Session Work" is ~30 lines of dense run-on prose ("Java, C++, and Rust WebSocket adapters track activity when frames arrive...", "Register child agents before running the parent: add_child_agent(namespace, name, child) in Python/C++, AddChildAgent in Go, addChildAgent in Java...") much of which is out of scope for a Rust flow skill. This fits 'mostly efficient but includes some unnecessary explanation or could be tightened' rather than anchor 2, since the excess is package-specific behavior poorly compressed, not generic padding.

3 / 5

Actionability

Concrete, syntactically complete Rust snippets cover the common cases ("let wf = axllm::flow(\"docs.coreFlow\").execute_with_options(...)", "parallel_flow.forward(&mut client, json!({...}))?", "forward_with_caching_function"), and the body points to runnable examples. It is not 5 because the Astra session section only names APIs ("with_child_agent in Rust", "set_tool_authorizer on C++ and Rust clients") without any Rust code example.

4 / 5

Workflow Clarity

As a patterns-reference skill, the progression is coherent — When To Use, Package Facts, Core Pattern, advanced patterns, run/stream/cache, then Guardrails — with explicit direction such as "Start from the complete programs under `examples/`" and "Start from package examples for exact native syntax before inventing a new call shape." It is not 5 because there are no explicit validation checkpoints or error-recovery guidance for the code the skill produces; it is above 3 because the learning sequence and entry points are clearly ordered.

4 / 5

Progressive Disclosure

References are clearly signaled in Package Facts ("Package API docs: `API.md` and `axir-api.json`", "Runnable examples: `examples/`") plus the external gallery link, but the entire "Astra Session Work" section is a dense wall of inline spec-level prose (MCP cancellation, WebSocket adapters, child-agent registration across five languages) that clearly belongs in a separate one-level-deep reference file. This matches 'content that should be separate is inline' rather than anchor 4, because that section is a substantial fraction of the body.

3 / 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, well-scoped description with an explicit 'Use when...' trigger, a clearly named niche package, and an enumerated list of covered subsystems. Its main weaknesses are the fused what/when structure and some jargon-heavy trigger terms that users may not naturally say.

DimensionReasoningScore

Specificity

The description names the concrete domain ("writing Rust code with `axllm`") and enumerates several specific subsystems ("flows, nodes, program graphs, nested programs, dynamic options, caching, and optimizer components"), matching the 'several specific items with minor gaps' anchor. It falls short of 5 because the only real action is 'writing' — capabilities are listed as topics rather than distinct concrete actions.

4 / 5

Completeness

The 'when' is explicit via the "Use when writing Rust code with `axllm` for flows..." trigger clause, and the 'what' is conveyed through the enumerated subsystems, satisfying both parts. It is not 5 because 'what' and 'when' are fused into a single clause — there is no separate, explicit statement of what the skill does.

4 / 5

Trigger Term Quality

Terms like "Rust", "axllm", "flows", and "caching" are natural phrases a user in this domain would say, giving good keyword coverage. It is not 5 because terms like "program graphs", "dynamic options", and "optimizer components" lean toward internal jargon and no synonyms or variations are offered.

4 / 5

Distinctiveness Conflict Risk

The "Rust code with `axllm`" pairing is a clear, narrow niche with distinct triggers, so risk of firing for the wrong skill is minimal. The description matches the top anchor exactly.

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.