Skip to content

Kartik-Doye/Ravengard-AI-Recruiter / .agents/skills/blueprint

blueprint

Verified

Documents observed or proposed software architecture with evidence-backed context, container, component, dynamic, data-flow, and deployment views. Use when the user needs architecture diagrams, onboarding material, trust boundaries, or a design view before a major change.

FileKindSize
SKILL.mdskill md13.2 KB
assets/template.htmlother5.3 KB
references/diagram-guide.mdmarkdown3.2 KB
references/research.mdmarkdown3.1 KB
scripts/validate-mermaidother3.4 KB

references/diagram-guide.md

View on GitHub
# Diagram menu + Mermaid patterns

## The menu: question → diagram

| The reader's question | Diagram | Mermaid type |
| --- | --- | --- |
| What is this system and what's around it? | System context (C4 L1) | `flowchart LR` |
| What are the major deployable pieces? | Containers (C4 L2) | `flowchart LR` with subgraphs |
| How is the code organized inside an app? | Components (C4 L3) | `flowchart TB` |
| What happens when a request comes in? | Request sequence | `sequenceDiagram` |
| How does data move and where does it live? | Data flow / ER | `flowchart TD` / `erDiagram` |
| What happens when X fails? | Failure path | `flowchart TD` (red edges) |
| How is it deployed and released? | Deployment / CI | `flowchart LR` grouped by env |
| Where are the trust boundaries / threat vectors? | Threat Model | `flowchart TD` with `subgraph` for zones |
| Who does what over time (async)? | Event flow | `sequenceDiagram` |

Pick by question, not by completeness. Delete any diagram that does not answer
a named reader question.

## Style rules (all diagrams)

- `%%{init: {'theme':'dark'}}%%` is unnecessary: the HTML template
  themes them; keep sources theme-neutral.
- Node labels: real names + a role hint: `api["API (Hono)"]`.
- Data stores: cylinder shape `[(Postgres)]`; external systems: box with
  dashed border for inferred, solid for verified.
- Split when labels or relationships stop being legible, then link the views
  with one sentence.
- **Explicit Trust Boundaries:** Use subgraphs (`subgraph TrustZone`) to demarcate where data crosses from public to private, or between different permission contexts. Security must not be an afterthought.
- Direction left-to-right for flows that follow time; top-down for structure.

## Patterns

### System context

```mermaid
flowchart LR
  user["User (browser)"]
  system["Attendance API (Hono)"]
  db[("SQLite")]
  mail["Resend (email)"]
  user -->|HTTPS| system
  system --> db
  system -->|magic-link email| mail
```

### Request sequence

```mermaid
sequenceDiagram
  participant B as Browser
  participant A as API
  participant D as DB
  B->>A: POST /login {email}
  A->>D: find user, create token
  D-->>A: token row
  A-->>B: 200 {ok}
  Note over A: rate-limited 5/min/IP
```

### Failure path

```mermaid
flowchart TD
  req["POST /checkin"] --> api["API"]
  api -->|db down| retry["queue retry (max 3)"]
  retry -->|still failing| dlq["dead-letter + alert"]
  api -->|ok| resp["200"]
```

### ER

```mermaid
erDiagram
  USER ||--o{ CHECKIN : makes
  USER { string id PK  string email  datetime created_at }
  CHECKIN { string id PK  string user_id FK  date day  time at }
```

## HTML deliverable

The bundled HTML template (in this skill's assets folder) is the shell:
sidebar nav, content sections,
light/dark toggle, zoom buttons per diagram, print CSS. Fill every
placeholder (`{{TITLE}}`, nav items, sections); delete sections you don't
use. Diagrams render via Mermaid from CDN: if the reader is offline the
template shows a notice pointing at the `.mmd` files (which is also why
the `.mmd` sources always ship alongside).

Print-friendly means: `@media print` in the template collapses the
sidebar, forces light theme, and avoids page-breaks inside diagrams,
verify with a print preview before shipping.
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 323kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall