Build a reviewable PCB fabrication candidate from engineering requirements

Why use this

1. A PCB can pass automated checks while a connector orientation, footprint or component substitution remains wrong for the intended assembly. Manufacturing files make these ambiguities expensive because a clean export can still describe an unusable board.

2. This job could organize a fabrication candidate around explicit engineering decisions and evidence. Keeping requirements, component provenance and unresolved geometry issues beside the files would help an engineer decide what is ready for review and what must remain blocked.

Why an agent / simpler alternative

Useful judgment

Cross-check structured requirements, pin mappings and design reports for engineering review.

Simpler alternative

Use a proven reference design when the required circuit already matches a supported configuration.

Unsuitable when

The design is safety-critical or the agent is expected to certify electrical performance.

Proposed inputs

1. Name

   Engineering requirements

   Owner

   Design owner

   Minimum scope

   Power, interfaces, mechanical constraints and intended operating conditions

   Quality check

   Identify critical missing specifications before component selection.

2. Name

   Design and component references

   Owner

   Engineer

   Minimum scope

   Authorized KiCad copies, datasheets, footprints and approved parts

   Quality check

   Verify exact part variants and pin-one/orientation conventions.

Access and setup — proposed or unknown

Proposed minimum

1. Work on versioned design copies and propose validation before local export or supplier interaction.

Compatibility unknowns

1. Unknown: KiCad versions, helper compatibility, part availability and manufacturing constraints.

External actions requiring approval

1. Installing tools, uploading designs, ordering boards or publishing intellectual property.

Expected deliverable and review checks

Proposed artifact

Fabrication-candidate package plan with requirement traceability, check reports, unresolved issues and cost-source fields.

Critical checks

1. Schematic pin assignments agree with footprints and connector orientation.

2. ERC/DRC passes do not override unresolved mechanical or component evidence.

Missing output evidence

No new run or output inspection in this editorial layer; see linked source-specific limits.

Failure signals and proposed responses

1. Signal

   A footprint matches the part name but reverses connector orientation.

   Why it matters

   Assembly may be physically incompatible despite electrical connectivity.

   Proposed response

   Inspect datasheet and board geometry together and block readiness until confirmed.

2. Signal

   A suggested replacement has a different voltage or package rating.

   Why it matters

   A sourcing change could invalidate the design.

   Proposed response

   Require engineer review of the substitution against all relevant requirements.

Effort and cost — unknown

Setup work

1. Complete requirements and verify component/footprint provenance.

Ongoing review

1. Review design-rule exceptions and reassess supplier substitutions.

Unknowns

1. Unknown: fabrication prices, component supply, engineering review and physical validation costs.

First check — proposed, unexecuted

Status

proposed_unexecuted

Synthetic fixture

A synthetic low-voltage connector board has a mirrored footprint and a substitute part with an incompatible rating.

Acceptance checks

1. Both issues block the fabrication candidate.

2. No supplier upload or order is implied by a clean file export.

Stop conditions

1. Critical requirements, orientation or component suitability remain unresolved.

Limits

1. A reviewable file package is not evidence of a working circuit, manufacturability or regulatory certification.