About paw-cra-quality-control
paw-cra-quality-control is a campaign-level quality workflow for creative teams that aggregates checks across visual and video assets. It reviews individual assets and cross-asset consistency, then produces a report with severity-rated issues and revision tickets.
Key features
- 01Scan campaign assets and manifests
- 02Check visuals, video, and brand rules
- 03Review cross-asset consistency
- 04Generate pass/fail reports
- 05Create and route revision tickets
Use cases
- 01Review campaign design and video assets
- 02Check brand consistency across asset sets
- 03Identify assets needing rework
- 04Run quality checks without interaction
Install
Copy the command and run it in a terminal.
npx skills add https:/ / github.com/ pawbytes/ skill-suites --skill paw-cra-quality-controlAfter installing, type/paw-cra-quality-controlin Claude Code to confirm it loaded.
Bundled files
References and scripts installed along with SKILL.md.
references/01-asset-collection.md02-per-asset-qc.md03-cross-asset-consistency.md04-report-and-route.mdDetails
From the project READMEQuality Control
Overview
This workflow aggregates quality control across all produced visual and video assets in a campaign, then generates a pass/fail report with revision tickets for anything that needs rework. Act as a meticulous QA lead who understands both brand standards and platform technical requirements.
Args: Accepts --headless / -H for autonomous execution against a campaign folder.
Output: qa-report.md with per-asset pass/fail verdicts, severity-rated issues, and revision tickets routed back to owning production workflows.
Key distinction: This is campaign-level cross-asset QC. Individual production workflows (Designer, Video Producer) already have embedded per-asset review gates. This workflow aggregates those results and adds cross-asset consistency checking that no single production workflow can do alone --- verifying that all campaign assets feel like a cohesive set.
On Activation
Load available config from {project-root}/.pawbytes/config/config.yaml and {project-root}/.pawbytes/config/config.user.yaml (root level and cra section). If config is missing, let the user know paw-cra-setup can configure the module at any time. Resolve and apply throughout the session (defaults in parens):
{user_name}(null) --- address the user by name{communication_language}(English) --- use for all communications{document_output_language}(English) --- use for generated document content{output_directory}(.pawbytes/creative-suites) --- base output path
Load shared agency memory from {project-root}/.pawbytes/creative-suites/index.md. Identify the active brand and campaign. Load brand guidelines from .pawbytes/creative-suites/brands/{brand-name}/guidelines.md.
Campaign folder resolution: Accept a campaign folder path as argument, or infer from the active campaign in memory. The campaign folder is expected at .pawbytes/creative-suites/brands/{brand-name}/campaigns/{campaign-name}/ or .pawbytes/creative-suites/campaigns/{campaign-name}/. If no campaign can be resolved, ask the user.
If --headless or -H is passed, run the full pipeline without interaction: collect assets, run all QC checks, generate report, route results.
Otherwise, greet the user, confirm the target campaign, and proceed through the stages below. Offer a summary of what will be checked before starting.
Stages
This workflow progresses through four stages. Each stage writes findings to the in-progress qa-report.md using the document-as-cache pattern for compaction survival.
| Stage | Purpose | Reference |
|---|---|---|
| 1. Asset Collection | Scan campaign folder, discover all manifests and assets | Load ./references/01-asset-collection.md |
| 2. Per-Asset QC | Visual checks, video checks, brand compliance per asset | Load ./references/02-per-asset-qc.md |
| 3. Cross-Asset Consistency | Verify cohesion across the full campaign asset set | Load ./references/03-cross-asset-consistency.md |
| 4. Report & Route | Generate final report, create revision tickets, update status | Load ./references/04-report-and-route.md |
Severity Model
All issues use a three-tier severity:
| Severity | Meaning | Effect |
|---|---|---|
| Critical | Asset is unusable or off-brand --- wrong dimensions, broken file, wrong logo, unreadable text | Asset fails. Revision ticket created. |
| Warning | Asset is usable but suboptimal --- slight color drift, audio slightly hot, missing optional metadata | Asset passes with note. Revision recommended. |
| Info | Observation or suggestion --- could improve but not a quality gate issue | Logged for reference. No revision ticket. |
An asset passes if it has zero critical issues. A campaign is QC-approved only when all assets pass.
Manifest Types
Production workflows emit these manifest files. This workflow reads all of them:
| Manifest | Producer | Contains |
|---|---|---|
design-manifest.json | Designer workflows | Image assets, dimensions, format, brand colors used |
video-manifest.json | Video Producer workflows | Video assets, codec, resolution, duration, subtitle status |
clip-manifest.json | paw-cra-video-clips | Extracted clips with reframing and platform targets |
batch-manifest.json | paw-cra-design-batch | Batch-produced visual assets with content calendar mapping |