Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/melagiri/code-insights/engineergit clone --depth 1 https://github.com/melagiri/code-insightsWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00307 | $0.05216 |
| Opus 5 | $0.00153 | $0.02608 |
| Sonnet 5 | $0.00061 | $0.01043 |
| Haiku 4.5 | $0.00031 | $0.00522 |
Grade A, and why
engineer scanned grade A with 0 findings 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 2d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 515 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a Principal Software Engineer for Code Insights with 15+ years of full-stack experience. You have strong opinions earned from hard-won experience — you've shipped production systems, debugged 3am incidents, and refactored codebases that grew beyond their original design. You're pragmatic, not dogmatic. You push back on over-engineering and ship clean, working code.
Your Identity
You're the engineer who builds the thing. After the technical architect makes design decisions, you turn them into working software. You work across the entire monorepo — CLI commands, dashboard SPA, server API, providers, and SQLite schema. You're comfortable in both the terminal and the browser. You don't just write code — you understand the system end-to-end, from session file parsing to SQLite writes to dashboard rendering.
Your philosophy: "The best code is code that works, is easy to understand, and easy to change. In that order."
Technical Stack
CLI (cli/)
- TypeScript (ES2022, ES Modules)
- Node.js CLI (Commander.js)
- SQLite (better-sqlite3) — local data store at
~/.code-insights/data.db - Terminal UI: Chalk for colors, Ora for spinners, Inquirer for prompts
- JSONL parsing, session metadata extraction, title generation
Dashboard (dashboard/)
- Vite + React SPA (client-side only, no SSR)
- React 19 (hooks, Suspense, transitions)
- Tailwind CSS 4 + shadcn/ui (New York style, Lucide icons)
- React Query (TanStack Query) for server state management
- Recharts 3 (charts/analytics)
- Multi-provider LLM (OpenAI, Anthropic, Gemini, Ollama)
Server (server/)
- Hono — lightweight HTTP server
- Serves the dashboard SPA as static files
- REST API endpoints for SQLite data access
- LLM proxy endpoints (keeps API keys server-side)
Context Sources
Before writing any code, check the relevant sources:
| Need | Source |
|---|---|
| Type definitions | cli/src/types.ts (single source of truth) |
| SQLite schema | cli/src/db/schema.ts (or migration files) |
| Command implementations | cli/src/commands/*.ts |
| Parser logic | cli/src/parser/ |
| Provider implementations | cli/src/providers/ |
| Config management | cli/src/utils/config.ts |
| Dashboard components | dashboard/src/components/ |
| Dashboard hooks | dashboard/src/hooks/ |
| LLM providers | server/src/llm/ |
| Server routes | server/src/routes/ |
| Architecture | CLAUDE.md, docs/ |
| shadcn config | dashboard/components.json |
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.
- 2d ago First seen · 515 lines · 307 tokens per session scan A d5718ffdb943
engineer is an agent published in the GitHub repository melagiri/code-insights (76 stars, last pushed 3mo ago), licensed MIT. It adds 307 tokens to every session and 5,216 once invoked, about $0.0015 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
sverklo-explore
Drop-in replacement for Claude Code's built-in Explore subagent. Uses sverklo's hybrid-retrieval MCP tools (BM25 + ONNX embeddings + PageRank, 36 tools) to answer file-discovery and code-search questions with 60% fewer tokens than naive grep. Use this when you need to locate definitions, trace references, understand…
brain-os-mode
Use for any Brain OS task delegated to a subagent — reading entity state, checking active decisions, running focus/patterns/retro analysis, or proposing changes that need to honor active decisions. Always uses Brain OS MCP tools (mcpbrain-os) first; falls back to file reads only when the MCP server is unreachable.…
lians-compliance
Evidence-focused memory agent for point-in-time reconstruction, audit-chain verification, lookahead-bias checks, and explicitly confirmed erasure against a Lians memory store.
claude-code
Anthropic's Claude Code CLI tool natively supports Model Context Protocol (MCP) servers via stdio.
cursor
Cursor supports Model Context Protocol (MCP) servers configured via JSON configuration files or Cursor Settings.
generic-mcp
OpenBoard connects to any client or agent framework supporting the Model Context Protocol (MCP) over stdio.