app-user-docs

Generate comprehensive user perspective documentation for any app by running a two-phase subagent workflow — Phase 1 surveys the codebase and produces a structured inventory, Phase 2 writes the final

SKILL.md

GitHub

App User Docs Skill

Generates a comprehensive user-perspective document for any application using a two-phase subagent workflow. The output is an honest, exhaustive markdown document written from the user's point of view — not developer docs, not a technical spec.


Step 1: Fill Out the App Profile

Before spawning any subagents, confirm or fill in this profile. Extract what you can from the conversation context — only ask the user about genuine gaps.

APP PROFILE
-----------
App name:
One-line description:
App type: [voice-first | standard-web | iot-embedded | hub-integrations | mixed]
Primary interface: [voice | click-based | both | physical/sensor]
Known integrations (external systems):
Core workflow or pipeline (if any):
Known incomplete features (if any):
Who is this doc for: [personal reference | handoff | contractor docs | formal manual]
Tone: [operational/blunt | professional | beginner-friendly]

App type determines which variant instructions Phase 2 uses. See references/app-type-variants.md.


Step 2: Phase 1 — Codebase Surveyor Subagent

Spawn a subagent with the Phase 1 prompt from references/phase1-surveyor.md.

Inject the completed App Profile at the top of the prompt before dispatching.

Phase 1 output is a raw structured inventory — no prose, no interpretation. It produces:

  • All UI panels/screens found, with file references
  • All user-facing actions (clicks, inputs, triggers)
  • All voice command handlers (if app type is voice-first or mixed)
  • All integration adapters and what they connect to
  • All data flows (in and out)
  • All TODOs, stubs, feature flags, and incomplete implementations
  • File path references for every item found

Before proceeding to Phase 2, present the inventory to the user and ask:

"Does this inventory look complete? Any areas that seem thin or missing?"

If the user identifies gaps, re-run Phase 1 with targeted directory hints before continuing.


Step 3: Phase 2 — Document Writer Subagent

Spawn a subagent with the Phase 2 prompt from references/phase2-writer.md.

Inject at the top:

  1. The completed App Profile
  2. The full Phase 1 inventory output
  3. The relevant app-type variant section from references/app-type-variants.md

Phase 2 output is the final user-perspective markdown document.


Step 4: Deliver

Present the markdown document to the user. Ask:

"Does this capture everything accurately? Any sections that need more depth or correction?"

Iterate on specific sections as needed — no need to re-run the full workflow for minor corrections.


Status Flags

Use these consistently throughout all generated documents:

  • ✅ Working — confirmed connected and functional
  • ⚠️ Partial — exists but incomplete or intermittently functional
  • ❌ Not connected — UI exists but no live backend/integration
  • ❓ Unconfirmed — found in code but UI state unknown

Reference Files

Read these when needed — do not load all of them upfront:

File When to read
references/phase1-surveyor.md When ready to dispatch Phase 1 subagent
references/phase2-writer.md When ready to dispatch Phase 2 subagent
references/app-type-variants.md When preparing Phase 2 prompt — inject relevant variant section only