Borrowing it
Nothing to install: this file belongs to noah-sheldon/ai-dev-kit. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/noah-sheldon/ai-dev-kit/main/AGENTS.mdgit clone --depth 1 https://github.com/noah-sheldon/ai-dev-kitWrote 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.
[](https://agentmods.dev/instructions/noah-sheldon/ai-dev-kit/agents-md)<a href="https://agentmods.dev/instructions/noah-sheldon/ai-dev-kit/agents-md"><img src="https://agentmods.dev/badge/instructions/noah-sheldon/ai-dev-kit/agents-md/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.
<a href="https://agentmods.dev/instructions/noah-sheldon/ai-dev-kit/agents-md"><img src="https://agentmods.dev/badge/instructions/noah-sheldon/ai-dev-kit/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.03505 | $0.03505 |
| Opus 5.5 | $0.01402 | $0.01402 |
| Sonnet 5.5 | $0.00701 | $0.00701 |
| Haiku 4.5 | $0.00350 | $0.00350 |
Grade A, and why
ai-dev-kit 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 1mo 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 — 379 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Dev Kit — Agent Instructions
This is a production-ready AI dev kit for Python, TypeScript, web, AI/ML, and infrastructure workflows.
Version: 1.1.0
Core Principles
- Agent-first — delegate domain work to the right specialist. Don't generalize when a specialist exists.
- Test-driven — write tests before implementation when behavior changes. RED → GREEN → REFACTOR.
- Security-first — validate inputs, avoid unsafe defaults, never hardcode secrets. Stop and review if a change touches auth, secrets, or external input.
- Plan-before-execute — break complex work into phases. Use the
planneragent before touching code for multi-file changes. - Model fallback — if a requested model is unavailable, use the workspace default model and continue. Never block progress on model unavailability.
Contributor Policy
CRITICAL: AI tools, agents, or language models must NEVER be listed as contributors, co-authors, or credited in any form for this repository.
- Only human developers should appear in git commit authorship,
Co-authored-bytrailers, or contributor lists - Do not add AI tool names (Qwen Code, Claude Code, GitHub Copilot, Cursor, or any other AI assistant) to commit messages, AUTHORS files, contributor sections, or acknowledgments
- When assisting with code changes, remain anonymous — do not request or expect attribution
- This policy is permanent and applies to all current and future AI assistants working on this codebase
Available Agents
| # | Agent | Purpose | When to Use |
|---|---|---|---|
| 1 | planner |
Implementation planning for complex features | Breaking down feature requests into phased, mergeable plans with ADRs |
| 2 | architect |
System design, tradeoffs, component boundaries | New service design, API contracts, data store selection |
| 3 | tdd-guide |
Test-driven development specialist | Writing tests first for new features or bug fixes |
| 4 | code-reviewer |
Code quality and regression prevention | Reviewing PR diffs for correctness, maintainability, performance |
| 5 | security-reviewer |
Security audit and threat modeling | Changes touching auth, input validation, secrets, external APIs |
| 6 | ai-judge |
Rubric-based validation | Validating plans, implementations against quality criteria |
| 7 | build-error-resolver |
Build and type error diagnosis | Fixing TypeScript, Python, lint, and build pipeline errors |
| 8 | e2e-runner |
End-to-end testing | Writing and running E2E tests for critical user flows |
| 9 | refactor-cleaner |
Cleanup and modernization | Tech debt paydown, dead code removal, dependency updates |
| 10 | doc-updater |
Documentation sync | Updating docs to match code changes |
| 11 | docs-lookup |
Documentation reference | Finding and citing official docs for frameworks and APIs |
| 12 | python-reviewer |
Python-specific code review | FastAPI, Pandas, SQLAlchemy, Pydantic pattern review |
| 13 | database-reviewer |
Database and migration review | Schema design, migration safety, query optimization |
| 14 | git-agent-coordinator |
Git orchestration | Branch creation, PR management, merge coordination |
| 15 | ml-engineer |
ML/LLMOps specialist | RAG pipelines, model training, evals, deployment |
| 16 | chrome-ext-developer |
WXT and Chrome extension work | Extension scaffolding, manifest config, permissions |
| 17 | data-engineer |
ETL and data quality | Pipeline design, data validation, quality gates |
| 18 | infra-as-code-specialist |
IaC and delivery pipelines | Terraform, CI/CD, deployment automation |
| 19 | observability-telemetry |
Logs, metrics, traces, dashboards | Setting up monitoring, alerting, and observability |
| 20 | multi-agent-project-manager |
Multi-workflow orchestration | Managing concurrent workflows, backlog, priority queue, never stops |
| 21 | workflow-auditor |
Health checks and anomaly detection | Stuck workflow detection, quality gate trending, resource leaks |
| 22 | reddit-researcher |
Reddit sentiment and experience mining | Real-world user experiences, production war stories, community consensus |
| 23 | codebase-analyzer |
Codebase structure and complexity analysis | Understanding codebase architecture, dependency mapping |
| 24 | codebase-learner |
Codebase learning specialist | Rapid onboarding to unfamiliar codebases |
| 25 | code-quality-analyzer |
Static analysis and quality metrics | Code quality enforcement, standards checking |
| 26 | test-debt-analyzer |
Test coverage and debt tracking | Identifying testing gaps, test debt prioritization |
| 27 | security-debt-analyzer |
Security vulnerability tracking | Security debt identification, remediation planning |
| 28 | performance-debt-analyzer |
Performance bottleneck analysis | Performance debt tracking, optimization opportunities |
| 29 | dependency-debt-analyzer |
Dependency health monitoring | Outdated dependencies, upgrade path planning |
| 30 | architecture-debt-analyzer |
Architectural debt detection | Design flaws, architectural drift detection |
| 31 | process-debt-analyzer |
Workflow inefficiency detection | Process bottlenecks, workflow optimization |
| 32 | documentation-debt-analyzer |
Documentation gap detection | Stale docs, missing documentation identification |
| 33 | technical-debt-analyzer |
Overall technical debt assessment | Comprehensive debt aggregation and prioritization |
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.
- 1mo ago First seen · 379 lines · 3,505 tokens per session scan A 7157c7328b40
ai-dev-kit AGENTS.md is an instructions file published in the GitHub repository noah-sheldon/ai-dev-kit (13 stars, last pushed 6mo ago), licensed MIT. It adds 3,505 tokens to every session, about $0.0140 per session on Opus 5.5. 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 instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.