agentleFS
timscheueraitimscheuerai/gtm-architecture-template/.agents/skills/oxygen-workflow-authoring/SKILL.md

oxygen-workflow-authoring

Skill2 starsChanged 2 months agoMITView on GitHub

Author, validate and operate deterministic OXYGEN Workflows with explicit inputs, hosted runs, current schemas and inspectable effects.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read composition, the runtime contract and the relevant function definitions.

    line 8
  2. Distinctive:

    Use Workflows for deterministic control flow, Tables for row dependencies, Sequences for outreach cadence and Agents for adaptive goals.

    line 10
  3. Distinctive:

    Search existing definitions/templates and resolve resource IDs from the chosen workspace.

    line 18

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Compose the hosted workflow

Read composition, the runtime contract and the relevant function definitions. This kit already contains six disabled workflow graphs.

Use Workflows for deterministic control flow, Tables for row dependencies, Sequences for outreach cadence and Agents for adaptive goals. Local files author definitions; OXYGEN executes production runs.

oxygen commands get "workflows schema" --json
oxygen workflows schema --subject graph --json
oxygen capabilities search "workflow trigger from a qualified account" --json
  1. Search existing definitions/templates and resolve resource IDs from the chosen workspace.
  2. Inspect the selected trigger's current event schema, including source, event and filters. Do not equate a human event label with an exact trigger.
  3. Author typed inputs, closed output schemas, explicit dependencies and error/hold paths. Avoid # in node IDs; it is reserved for iteration IDs.
  4. Lint for the intended draft/publish phase, then apply within the requested scope. Inspect the saved version and trigger status.
  5. Discover the current call preview and dry-run grammar. A call preview describes effects; a dry run may perform internal reads or permitted GETs. Inspect both before live scope.
  6. Execute or enable only the authorized rows/events, effects and credit ceiling. Tail the durable run and inspect outputs, retries and costs.

Code nodes use declared inputs. For this kit, edit the pure source/contract and run npm run build; the generated (inputs) => ... expression is the deployed source. Do not insert ambient credentials, clocks or network calls.

An outbound graph enrolls through the native Sequence action. Inspect its current state and enrollment semantics; pending enrollment and active delivery differ. It must not bypass sender, suppression or reply-stop controls with raw provider sends.

Return definition/version, lint results, previewed effects, run link and verified state. Reconcile ambiguous external effects before retrying; replay and one-node tests must use the current server-supported semantics.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.