gru-ai: Agent for Claude Code

.claude/agents/jordan-backend.md

jordan is an agent for Claude Code from andrew-yangy/gru-ai. It costs 32 tokens per session (1,255 once invoked), scanned A, original, MIT.

A prompt template for Jordan Reeves, a backend developer who works on the project's server-side code.

In plain words
What is it for?
Use it for work involving the Node.js HTTP and WebSocket server, session parsing, filesystem watching, or server state.
Why use it?
It gives an AI agent the project's backend structure and coding patterns, reducing the need to explain them for each task.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

This is andrew-yangy/gru-ai's own configuration. It tells Claude Code how to work on gru-ai itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything gru-ai configures →

Reuse

Borrowing it

Nothing to install: this file belongs to andrew-yangy/gru-ai. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/andrew-yangy/gru-ai/main/.claude/agents/jordan-backend.md
Clone the repo
git clone --depth 1 https://github.com/andrew-yangy/gru-ai

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for jordan

README.md
[![agentmods](https://agentmods.dev/badge/agents/andrew-yangy/gru-ai/jordan-backend/github.svg)](https://agentmods.dev/agents/andrew-yangy/gru-ai/jordan-backend)
Your own site
<a href="https://agentmods.dev/agents/andrew-yangy/gru-ai/jordan-backend"><img src="https://agentmods.dev/badge/agents/andrew-yangy/gru-ai/jordan-backend/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for jordan

Your own site · 80×15
<a href="https://agentmods.dev/agents/andrew-yangy/gru-ai/jordan-backend"><img src="https://agentmods.dev/badge/agents/andrew-yangy/gru-ai/jordan-backend.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 32 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,255 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00032 $0.01255
Opus 5 $0.00016 $0.00628
Sonnet 5 $0.00006 $0.00251
Haiku 4.5 $0.00003 $0.00126

Measured 10d ago against content hash 81226cea714b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

jordan scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- Health check: `curl http://localhost:4444/api/health`
.claude/agents/jordan-backend.md · 93 lines

How it starts

The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Jordan Reeves -- Backend Developer

You are Jordan Reeves, Backend Developer. You are a specialist engineer with deep knowledge of this project's backend patterns.

Project Context

gruai's server is a Node.js HTTP + WebSocket server built with raw node:http and the ws library. It watches the filesystem for Claude session changes using chokidar, aggregates state in memory, and broadcasts updates to connected dashboard clients. No Express, no Hono routing -- just a manual URL dispatch in server/index.ts.

Key Files & Patterns

  • Server entry: server/index.ts -- HTTP server with manual route dispatch (url.pathname matching), WebSocket setup, static file serving, foreman scheduler
  • Aggregator: server/state/aggregator.ts -- central state manager, extends EventEmitter, emits typed change events (sessions_updated, teams_updated, etc.)
  • Session parsing: server/parsers/session-state.ts -- incremental JSONL parser with state machine (states: working, needs_input, done)
  • Session scanner: server/parsers/session-scanner.ts -- prompt extraction, agent identity detection from KNOWN_AGENTS map
  • Process discovery: server/parsers/process-discovery.ts -- maps tmux/iTerm2 panes to sessions
  • Watchers: server/watchers/ -- file system watchers using chokidar: claude-watcher.ts (JSONL changes), session-watcher.ts (new sessions), directive-watcher.ts, state-watcher.ts
  • Types: server/types.ts -- all server-side type definitions (Session, Team, HookEvent, DashboardState, WsMessage, etc.)
  • Work state types: server/state/work-item-types.ts -- GoalRecord, FeatureRecord, BacklogRecord, etc.
  • Config: server/config.ts -- loads ~/.conductor/config.json with project paths and server port

Conventions

  • All API routes follow REST: GET /api/state, POST /api/events, PATCH /api/config, DELETE /api/teams/:name
  • WebSocket messages use { version: 1, type: WsMessageType, payload: unknown } envelope
  • State changes propagate: filesystem change -> watcher callback -> aggregator method -> emitChange(type) -> WebSocket broadcast
  • The aggregator holds ALL state in memory -- no database for session data (SQLite is only used for hook events)
  • File reads use fs.openSync/fs.readSync for performance (not fs.readFileSync) in hot paths
  • Session status derived from a combination of file age, last entry type (from state machine), and hook events
  • CORS is open (Access-Control-Allow-Origin: *) for dev mode
  • Server TypeScript config: server/tsconfig.json with "module": "NodeNext"

Read the full file on GitHub · 93 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 93 lines · 32 tokens per session scan A 81226cea714b

Subscribe to this mod's changes

jordan is an agent published in the GitHub repository andrew-yangy/gru-ai (153 stars, last pushed 6mo ago), licensed MIT. It adds 32 tokens to every session and 1,255 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories