Use when writing Java code with `dev.axllm:ax` for flows, nodes, program graphs, nested programs, dynamic options, caching, and optimizer components.
66
80%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Passed
No findings from the security scan
Fix and improve this skill with Tessl
tessl review fix ./packages/java/skills/ax-java-flow/SKILL.mdThis skill helps an agent write Java code with the generated Ax package dev.axllm:ax. Use the generated package API, examples, and manifests; do not import TypeScript-only APIs unless you are editing the TypeScript package.
dev.axllm:ax.API.md and axir-api.json.axir-capabilities.json.examples/.javascript-quickjs, python-pyodide.AxGen draft = Ax.ax("topicText:string -> draftText:string");
AxFlow wf = Ax.flow(java.util.Map.of("id", "docs.coreFlow"))
.execute("draft", draft, java.util.Map.of(
"reads", java.util.List.of("topicText"),
"writes", java.util.List.of("draftResult", "draftText")))
.returns(java.util.Map.of("draftText", "draftText"));Build each flow node from its own input/output contract.
AxGen classifier = Ax.ax("requestText:string -> route:class \"support, sales, engineering\"");
AxGen responder = Ax.ax("requestText:string, route:string -> responseText:string");Declare reads and writes so the responder waits for the typed route.
AxFlow branchFlow = Ax.flow(Map.of("id", "docs.branchFlow"))
.execute("classifier", classifier, Map.of("reads", List.of("requestText"), "writes", List.of("classifierResult", "route")))
.execute("responder", responder, Map.of("reads", List.of("requestText", "route"), "writes", List.of("responderResult", "responseText")))
.returns(Map.of("route", "route", "responseText", "responseText"));Independent reads let research and audience analysis share one planner group.
AxFlow parallelFlow = Ax.flow(Map.of("id", "docs.parallelFlow"))
.execute("research", research, Map.of("reads", List.of("topicText"), "writes", List.of("researchResult", "factList")))
.execute("audience", audience, Map.of("reads", List.of("topicText"), "writes", List.of("audienceResult", "audienceAngle")))
.execute("join", join, Map.of("reads", List.of("factList", "audienceAngle"), "writes", List.of("joinResult", "briefText")))
.returns(Map.of("briefText", "briefText"));A linear refinement pipeline makes each dependency explicit.
AxFlow refineFlow = Ax.flow(Map.of("id", "docs.refineFlow"))
.execute("draft", draft, Map.of("reads", List.of("topicText"), "writes", List.of("draftResult", "draftText")))
.execute("critique", critique, Map.of("reads", List.of("draftText"), "writes", List.of("critiqueResult", "critiqueText")))
.execute("revise", revise, Map.of("reads", List.of("draftText", "critiqueText"), "writes", List.of("reviseResult", "revisedText")))
.returns(Map.of("revisedText", "revisedText"));Forward accepts the provider client and the public flow inputs.
var output = parallelFlow.forward(client, Map.of("topicText", "Typed LLM workflows"));Start from the complete programs under examples/, then browse the larger gallery at https://axllm.dev/java/subsystems/flow/.
Ax.flow, AxFlowprovider-api examples only when the user explicitly has provider credentials available.no-key examples for deterministic local checks and provider request mapping.tools/*/skills/ into user packages.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.