100-core

100-core is a cursor rule for coding agents from d-padmanabhan/agent-engineering-handbook. It costs 854 tokens per session, scanned A, original, MIT.

A set of general engineering rules for solving the requested problem, making minimal changes, preserving existing behavior, and verifying the result.

In plain words
What is it for?
Use it to guide repository work, confirm requirements from evidence, avoid unrelated refactors, document important assumptions, and apply correctness and completeness checks.
Why use it?
It reduces unnecessary changes and catches incomplete, unsupported, or untested implementations.

Cursor rule

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 rules/d-padmanabhan/agent-engineering-handbook/100-core
Clone the repo
git clone --depth 1 https://github.com/d-padmanabhan/agent-engineering-handbook

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 100-core

README.md
[![agentmods](https://agentmods.dev/badge/rules/d-padmanabhan/agent-engineering-handbook/100-core.svg)](https://agentmods.dev/rules/d-padmanabhan/agent-engineering-handbook/100-core)
Your own site
<a href="https://agentmods.dev/rules/d-padmanabhan/agent-engineering-handbook/100-core"><img src="https://agentmods.dev/badge/rules/d-padmanabhan/agent-engineering-handbook/100-core.svg" alt="Measured on agentmods" height="20"></a>
Per session 854 This file is loaded in full into every session.
When invoked 854 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 $0.00854 $0.00854
Opus 5 $0.00427 $0.00427
Sonnet 5 $0.00171 $0.00171
Haiku 4.5 $0.00085 $0.00085

Measured 4d ago against content hash 4ddf793f6a4e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

100-core 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 4d 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.

rules/100-core.mdc · 60 lines

How it starts

The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Core Engineering Gates

This rule owns universal engineering policy only. Use the core engineering skill (${HANDBOOK_ROOT}/skills/core-engineering/SKILL.md) and the relevant domain skill for procedures, examples, and language-specific guidance.

Solve the Actual Problem

  • Confirm the requested outcome and constraints from evidence before changing code.
  • Make the smallest coherent change that solves the problem. Do not refactor unrelated modules, alter APIs, add dependencies, or introduce extension points without need.
  • Preserve existing behavior unless the request explicitly changes it.
  • State assumptions when they materially affect correctness; do not invent repository, runtime, vendor, or operational details.

Produce Complete, Maintainable Work

  • Deliver complete implementations within scope. Do not leave placeholders, TODO-only stubs, omitted branches, or examples that cannot run as presented.
  • Use precise domain names, cohesive units, understandable control flow, and explicit side effects, ownership, and resource lifecycles.
  • Reduce meaningful duplication without creating premature abstractions or hidden coupling.
  • Delete dead or misleading code when removal is in scope and behavior is preserved.
  • Comments explain intent, constraints, or non-obvious decisions; they do not restate syntax.

Correctness and Failure Behavior

  • Validate inputs at trust boundaries and fail with specific, safe, actionable errors.
  • Handle expected failures explicitly; preserve causal context and clean up acquired resources.
  • Bound network, subprocess, retry, queue, and background work with timeouts, cancellation, and lifecycle ownership.
  • Protect concurrency invariants and avoid unsynchronized shared mutable state.
  • Add enough structured observability to diagnose key operations and failures without logging secrets or unnecessary sensitive data.

Evidence and Current Information

  • Ground claims in code, tests, diagnostics, command output, measurements, or authoritative documentation.
  • For vendor, cloud, SDK, API, CLI, IaC provider, runtime, security, or authentication behavior, verify current official documentation, installed-tool help, schemas, or release notes before recommending syntax, defaults, versions, limits, or architecture.
  • Label legacy guidance and explain when it still applies.
  • Performance changes require a bottleneck hypothesis, expected metric impact, and a measurement plan. Do not optimize by intuition alone.

Read the full file on GitHub · 60 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. 4d ago First seen · 60 lines · 854 tokens per session scan A 4ddf793f6a4e

Subscribe to this mod's changes

100-core is a cursor rule published in the GitHub repository d-padmanabhan/agent-engineering-handbook (16 stars, last pushed 4d ago), licensed MIT. It adds 854 tokens to every session, about $0.0043 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.