mental-map

star 0

Build mental maps of application architecture by analyzing Caido MCP proxy traffic, grouping requests into flows like auth, cart, checkout, signup, login, forgot-password, and user-profile, then documenting sequence diagrams and replication notes

ghostonbutterbread By ghostonbutterbread schedule Updated 4/7/2026

name: mental-map description: Build mental maps of application architecture by analyzing Caido MCP proxy traffic, grouping requests into flows like auth, cart, checkout, signup, login, forgot-password, and user-profile, then documenting sequence diagrams and replication notes

Mental Map Analysis

Build mental maps of application architecture from Caido MCP proxy traffic.

Required Preflight

Read shared state in this order before mapping flows:

  1. notes/summary.md
  2. notes/observations.md
  3. checklist.md (auth, workflow, and business-logic items only)
  4. todo.md (workflow mapping or prerequisite items only)

Primary Analysis Surface

There is no dedicated agents/mental_map.py harness in this repo. Use Caido MCP proxy traffic as the source of truth, set the browser or replay client proxy to KAIDO_MCP_PROXY_URL, then classify captured requests into application flows.

What To Map

Prioritize end-to-end flows another agent would need to replay safely:

  • auth
  • signup
  • login
  • forgot-password
  • user-profile
  • cart
  • checkout
  • Any custom billing, admin, search, upload, or API workflow that materially changes state

Files

  • Playbook: $HARNESS_ROOT/prompts/mental-map-playbook.md
  • Output Root: $HARNESS_SHARED_BASE/{program}/agent_shared/application-structure/
  • Flow Template: $HARNESS_ROOT/agent_shared/templates/application-structure/flow-template.md

Output Contract

Write one markdown file per flow to:

$HARNESS_SHARED_BASE/{program}/agent_shared/application-structure/{flow-type}/{flow-name}.md

Each flow file must include:

  • Domain
  • Endpoints involved
  • Request sequence
  • Auth requirements
  • Session handling and CSRF notes
  • Data model
  • State transitions
  • Replication notes for another agent

Workflow

  1. Complete the required preflight reads in shared state order.
  2. Read prompts/mental-map-playbook.md.
  3. Connect the browser or replay client to KAIDO_MCP_PROXY_URL and capture the real workflow.
  4. Group requests into a concrete flow with entry points, dependencies, and state-changing operations.
  5. Write the diagram and structured notes to agent_shared/application-structure/{flow-type}/{flow-name}.md.
  6. Update notes/summary.md, notes/observations.md, and todo.md when the map exposes new testing lanes or prerequisites.
Install via CLI
npx skills add https://github.com/ghostonbutterbread/bug-bounty-harness --skill mental-map
Repository Details
star Stars 0
call_split Forks 0
navigation Branch main
article Path SKILL.md
More from Creator
ghostonbutterbread
ghostonbutterbread Explore all skills →