Works with the AI Agents you already use
code map
Generate high-fidelity Markdown code maps and Mermaid architecture diagrams directly from source code.
Free
code map
Example session with this skill installed
code map ./services/order-processor --artifact
- Read your context and instructions
- Compiled the code map
I have analyzed the order-processor repository. The inventory identified a Kotlin/Spring Boot stack with 12 endpoints and 4 event listeners.
Order Processor — Code Map of the Checkout Process
What it does for the business:
When a customer completes a checkout, this service validates inventory, reserves stock, and initiates the payment flow. It handles partial stock scenarios and ensures that failed payments trigger an inventory release.
Path root: src/main/kotlin/com/acme/orders/
High Level Architecture
flowchart LR
WebGateway["Web Gateway<br/>POST /v1/orders"] --> OP["Order Processor"]
OP --> DB[("Postgres<br/>orders_db")]
OP --> PaymentSvc["Payment Service<br/>(External)"]
OP -.-> InventorySvc["Inventory Service<br/>(Dashed: Unverified)"]
Key Flows
Flow: Successful Order Placement
sequenceDiagram
participant OC as OrderController
participant OS as OrderService
participant ER as EntityRepository
OC->>OS: placeOrder(dto)
OS->>ER: save(PendingOrder)
OS->>OS: validateStock()
Note right of OS: Check src/services/StockValidator.kt:45
OS->>OC: 201 Created
Tech Stack
| Technology | Version | Purpose |
|---|---|---|
| Kotlin | 1.9.22 | Primary Language |
| Spring Boot | 3.2.0 | Application Framework |
| PostgreSQL | 15 | Persistent Storage |
Risks & Known Issues
Swallowed Exception: OrderService.kt:88 catches and logs PaymentException without rethrowing or triggering a retry.
- TODO Cluster:
StockValidator.ktcontains 4 TODOs regarding multi-warehouse logic.
⚠️ Not determined from these sources
Gap Why it is open Where the answer likely lives Inventory Reservation Call leaves via InventoryClientinventory-servicerepo
Artifact Published: https://claude.site/artifacts/order-processor-map-7788
Next steps
- Commit the map to
docs/code-map.mdto track changes in PRs. - Run with
--repo inventory-serviceto bridge the inventory reservation gap. - Review the swallowed exception in
OrderService.kt:88.
Run cost
| Model | Input Tokens | Output Tokens | Cost (USD) |
|---|---|---|---|
| Claude 3.5 Sonnet | 4,250 | 880 | $0.026 |
| Total | $0.026 |
Estimates based on standard API pricing.
Connects securely to your tools. The creator never sees your data.
About this skill
Onboarding a new developer to a complex codebase is slow. Reading raw source code provides detail but lacks context, while high-level design documents often drift from reality. This skill bridges that gap by generating a technical map directly from your repository source.
What it does
- Architectural visualization generates Mermaid flowcharts showing inbound callers, outbound dependencies, and data stores.
- Traceable sequence diagrams maps 2-3 critical end-to-end business flows with real file-and-line citations.
- Entity relationship mapping builds ERDs from database entities and migrations found in the source.
- Module responsibility indexing creates a directory-to-purpose table so developers know where to start.
- Risk surfacing identifies framework pitfalls, TODO clusters, and swallowed exceptions.
How it works
- Targeting resolves a local path or git URL and performs a scripted inventory of tech stacks and entry points.
- Analysis uses a low-token exploration strategy to trace call paths from controllers to repositories without re-reading what the inventory already found.
- Generation writes a single
code-map.mdfile formatted for native rendering in GitHub, IntelliJ, and Confluence. - Validation identifies "Not determined" gaps where logic leaves the repo and offers to bridge them by scanning sibling checkouts or remote services.
Frameworks & tools
Works with any language or framework. Uses internal bash scripts for inventory, Mermaid.js for diagrams, and git for provenance. Optionally uses artifact-design for web-based publishing.
Why this beats prompting it yourself
Standard prompts often hallucinate architecture or miss subtle configuration-based dependencies. This skill uses a multi-step inventory process to ensure every claim traces back to a specific file and line number, preventing design drift.
Use cases
- Onboarding new hires to a legacy service by providing a map of key business flows.
- Pre-PR audits to visualize how a new feature changes the existing service architecture.
- Documentation debt recovery for services that lack up-to-date design specs.
- Cross-service discovery to see how a specific event or API call moves between multiple repositories.
Known limitations
Repository analysis is limited to what is accessible in the provided source or linked git remotes. Large-scale multi-repo mapping requires manual confirmation of sibling directory names.
Requires git and bash. Remote repo mapping needs network access. Agent limits apply based on codebase size to manage token costs. ERD generation depends on local entity or migration files.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 27 days ago
- Free with an Agensi account
Requires