agentsclimarketplace

Orchestrate technical design

Skill agentic-workflow-kit/technical-design/skills/orchestrate-technical-design

AI skills for right-sized technical design, review loops, and enforceable architecture boundaries.

Install
npx -y skills add agentic-workflow-kit/technical-design --skill orchestrate-technical-design

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use when a user wants to run the full technical design process hands-off or orchestrate part of it. Reads and applies inputs -> system-model -> structure -> author -> diagrams -> review loop -> enforce, uses the active DDD methodology profile, pauses for approvals/dispositions, stops at the requested boundary, and does not reimplement skill logic.

SKILL.md

3.1 KB, 570 tokens by cl100k_base, as published. Nobody here has run it

orchestrate-technical-design

Run the technical-design process as a composition-only runbook. Do not invent scope, methodology rules, or enforcement behavior. Read the sibling skill instructions and the active methodology profile at each stage.

References

  • Active methodology profile: ../../methodologies/ddd/README.md
  • Methodology profile contract: ../../docs/design/methodology-profile-contract.md
  • Lessons ledger: ../../docs/design/lessons-ledger.md

Flow

  1. Identify stop boundary. Supported boundaries: inputs, system-model, structure, author, diagrams, review, enforce.
  2. Inputs. Read and apply ../frame-technical-design/SKILL.md; produce the source map and InputResolution.
  3. System model. Continue frame-technical-design; produce AgreedSystemModel, architecture_mode, initial ddd_depth, blockers, and approval status. Stop if required approvals are pending.
  4. Structure. Read and apply ../author-technical-design/SKILL.md; produce DocStructurePlan and stop if structure approval is pending.
  5. Author. Continue author-technical-design; produce the DDD-first design and decisions.md only after approval gates are satisfied.
  6. Diagrams. Add diagrams only when useful, and only to explain approved entities, flows, lifecycles, or boundaries.
  7. Review loop.
    • Read and apply ../review-technical-design/SKILL.md.
    • Present the report and wait for user dispositions.
    • Record dispositions in decisions.md.
    • Apply accepted suggestions through author update mode.
    • Repeat until no blocking suggestion is open.
  8. Enforce. Read and apply ../enforce-architecture/SKILL.md; generate rules only from the settled enforcement map and prove seeded violations fail.

Stop conditions

Stop and report when:

  • the requested boundary has been reached;
  • InputResolution, AgreedSystemModel, or DocStructurePlan requires approval;
  • a blocking question remains unanswered;
  • diagrams would introduce architecture without a recorded decision;
  • a design or review tries to hand implementation scope decisions to a later executor;
  • an enforcement rule lacks a seeded violation;
  • the active methodology profile is missing required profile files.

Final summary

At every boundary, report:

  • artifact produced;
  • unresolved blockers;
  • assumptions needing approval;
  • whether user approval is required before continuing.

In the final summary, report the produced artifacts, open blockers, accepted/deferred/rejected decisions, enforcement status, and verification commands run.

What ships with it: 1 file

2.0 KB alongside SKILL.md

examples/

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.