Cognitive Ux Workflow
Design and audit SaaS applications, internal tools, and recurring knowledge-work workflows using evidence-informed UX. Use when people repeatedly create, review, organize, approve, or resolve work; return an object/action model, primary and recovery flows, state/undo rules, accessible interaction requirements, and time-to-correct-outcome validation.
Install
$ agentstack add skill-kfeng209-oss-cognitive-ux-cognitive-ux-workflow ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
Cognitive UX: Workflow SaaS
Scope
Use for product workflow surfaces. Use cognitive-ux-dashboard for data monitoring and cognitive-ux-admin for privileged administration.
Intake
- Identify user roles, objects, primary jobs, frequency, collaboration, and handoffs.
- Identify states, permissions, automation, irreversible outcomes, audit needs, and offline/error constraints.
- Identify existing mental models, device/input method, and success/failure measures.
State assumptions that could change the design. Define the primary task, its consequence of failure, and a measurable user outcome before proposing a layout.
Design procedure
- Model the product around user-recognizable objects, states, and actions; avoid exposing internal implementation vocabulary as the primary navigation.
- Map the happy path plus interruption, conflict, no-permission, empty, bulk, and recovery paths before choosing components.
- Keep system state visible; preserve context after filtering, sorting, navigation, async work, and automation.
- Use safe defaults, confirmation or undo for consequential changes, and specific error recovery without discarding work.
- Design automation as reviewable assistance with clear scope, status, provenance, and human override.
Read [archetype evidence](references/archetype-evidence.md). Label each non-obvious recommendation Standard / requirement, Evidence-informed default, or Hypothesis to test.
Required outputs
- object/action/state model
- primary, exception, and recovery flow maps
- permission, automation, and undo/confirmation specification
- component behavior and accessibility checklist
- time-to-correct-outcome study
- Evidence-to-decision table: decision → mechanism → source → label → limitation → test metric.
- Accessibility, ethics, and recovery-path checklist.
Guardrails
Do not hide state changes, make destructive actions irreversible without warning, use modal sequences as a substitute for flow design, or automate beyond user authority.
Related skills
Use cognitive-ux-dashboard, cognitive-ux-onboarding, and cognitive-ux-admin. Select the primary task before combining skills; add a secondary skill only when it changes the user outcome or risk.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: kfeng209-oss
- Source: kfeng209-oss/cognitive-ux
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.