coding-with-ai-lab AGENTS.md

Project instructions for a full-stack CRM, a system for managing companies, people, departments, addresses, activities, and sales opportunities.

In plain words
What is it for?
Use them when changing the Node.js, Angular, database, permissions, authentication, or automated task-processing code.
Why use it?
They give an AI coding assistant the application’s architecture, data model, authentication details, and agent-task workflow.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/atra-consulting/coding-with-ai-lab/agents-md
Clone the repo
git clone --depth 1 https://github.com/atra-consulting/coding-with-ai-lab

Made for: Codex, OpenCode.

Per session 2,245 This file is loaded in full into every session.
When invoked 2,245 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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 $0.02245 $0.02245
Opus 5 $0.01123 $0.01123
Sonnet 5 $0.00449 $0.00449
Haiku 4.5 $0.00225 $0.00225

Measured 2d ago against content hash 9a25f3f8e7cd, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

coding-with-ai-lab AGENTS.md 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.

AGENTS.md · 88 lines

How it starts

The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Project knowledge for any AI coding assistant. Claude-Code-specific configuration (subagents, spec reading lists) lives in CLAUDE.md, which imports this file.

Project

Full-stack CRM application. Node.js/TypeScript (Express + Drizzle ORM + libSQL/SQLite via @libsql/client) backend, Angular 21 frontend. German domain model: Firma, Person, Abteilung, Adresse, Aktivitaet, Chance. Local SQLite file at backend/data/crmdb.sqlite; production runs on Turso cloud when TURSO_DATABASE_URL is set. Authentication via hardcoded users (backend/src/config/users.ts, bcrypt-hashed passwords), session-based and persisted to the sessions table (LibsqlSessionStore). 3 users: admin/admin123 (ADMIN), user/test123 (USER), demo/demo1234 (ADMIN). Enforcement is role-based via requireRole('ADMIN'); users also carry a permissions array, but no requirePermission middleware is wired up.

Autonomous Agents (advanced workshop)

Two independent Claude-Code-in-CI agents. Both are documented in full in docs/specs/SPEC-API-TASKS.md.

  • Agent-task runner — drains the agent_task table (sources EMAIL, GITHUB_ISSUE, APP_LOG, ERROR_REPORT; lifecycle OPEN → IN_PROGRESS → DONE | REJECTED) via the /api/agent-tasks API, decides solve-or-reject, implements, and merges. Admin dashboard at /admin/agent-tasks. Workflow .github/workflows/agent-task-runner.yml, prompts .claude/prompts/agent-*.md.
  • GitHub-issue agent — works real GitHub issues labelled Refinement needed, one per run, triggered from the solve-github-issues card in /admin/cron. Implements-or-asks; opens a PR against main (never merged) or comments a question and adds Input needed. Status tracked on GitHub Project board #7. Workflow .github/workflows/github-issue-agent.yml, prompt .claude/prompts/agent-github-refinement.md.

Agent endpoints authenticate with requireAgentToken (AGENT_API_TOKEN); cron triggers with requireCronAuth (CRON_SECRET or admin session). See docs/specs/SPEC-API-TASKS.md for endpoint signatures, required secrets, and board mechanics.

Read the full file on GitHub · 88 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. 2d ago First seen · 88 lines · 2,245 tokens per session scan A 9a25f3f8e7cd

Subscribe to this mod's changes

coding-with-ai-lab AGENTS.md is an instructions file published in the GitHub repository atra-consulting/coding-with-ai-lab (5 stars, last pushed 7d ago), licensed MIT. It adds 2,245 tokens to every session, about $0.0112 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 2026-08-31.

Related

Other instructions, from other repositories