core: Instructions file for Claude Code

CLAUDE.md

core CLAUDE.md is an instructions file for Claude Code from asdecided/core. It costs 3,895 tokens per session, scanned A, a copy of core copilot-instructions.md, Apache-2.0.

A session guide for AsDecided, a project that stores settled decisions, roadmaps, and prompts for AI agents.

In plain words
What is it for?
Use it at the start of sessions or when handling releases, pull requests, refactoring, or other documented workflows.
Why use it?
It directs agents to the authoritative documents and helps them follow accepted project decisions.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md.

This is asdecided/core's own configuration. It tells Claude Code how to work on core 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 core configures →

Reuse

Borrowing it

Nothing to install: this file belongs to asdecided/core. 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/asdecided/core/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/asdecided/core

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 core CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/asdecided/core/claude-md/github.svg)](https://agentmods.dev/instructions/asdecided/core/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/asdecided/core/claude-md"><img src="https://agentmods.dev/badge/instructions/asdecided/core/claude-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.

agentmods 80×15 button for core CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/asdecided/core/claude-md"><img src="https://agentmods.dev/badge/instructions/asdecided/core/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 3,895 This file is loaded in full into every session.
When invoked 3,895 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin 94% copy Near-identical to another mod 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 $0.03895 $0.03895
Opus 5.5 $0.01558 $0.01558
Sonnet 5.5 $0.00779 $0.00779
Haiku 4.5 $0.00390 $0.00390

Measured 1mo ago against content hash a614728f0836, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

core CLAUDE.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.

Origin

This is a copy

94% identical to core copilot-instructions.md — 25 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

CLAUDE.md · 149 lines

How it starts

The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AsDecided — agent session context

This file is a router. Canonical agent guidance lives in decisions/prompts/, where the AsDecided corpus gates validate it. Do not add rules here — add them to the corpus artifact and they load through the imports below.

Loaded every session

@decisions/prompts/rac-agent-session-start.md @decisions/prompts/rac-agent-commit-guidelines.md

Situational prompts — read when the task calls for it, do not import

  • Pull request preparation: decisions/prompts/rac-agent-pr-guidelines.md
  • Minor release gate: decisions/prompts/rac-agent-release-gate-minor.md
  • Major release gate: decisions/prompts/rac-agent-release-gate-major.md
  • Refactoring and simplification: decisions/prompts/rac-agent-simplification-guidelines.md
  • Context compression: decisions/prompts/rac-agent-compression.md

Working corpus

  • Current roadmap: decisions/roadmaps/future/ (release execution is tracked in GitHub)
  • Historical roadmaps: decisions/roadmaps/archive/
  • Decisions (ADRs): decisions/decisions/

Settled decisions (AsDecided)

These decisions are already accepted. Do not re-open or contradict them; ask the AsDecided MCP tools (get_artifact, search_artifacts) for the full text before proposing a change that touches one.

  • RAC-01K8P7A3M5QZ — ADR-125: Distribute AsDecided as a Local Pilot App (Architecture)
  • RAC-01K8Q7MCP407 — ADR-129: Confine Rename Writes to the Corpus Root (Technical)
  • RAC-01K8Q7MCP408 — ADR-130: Transactional Rename Application (Technical)
  • RAC-01K8Q7MCP411 — ADR-128: Hard MCP Response Budgets (Technical)
  • RAC-01K8Q7MCP413 — ADR-127: Attributable MCP Audit Records (Technical)
  • RAC-01K8Q7MCP431 — ADR-131: Native Telemetry Is Local-Only (Product)
  • RAC-01K8Q7MCP432 — ADR-132: AsDecided MCP Server Identity (Product)
  • RAC-KTQ63DPSMF19 — ADR-001 Markdown First
  • RAC-KTQ63DPT6008 — ADR-002 AI Optional
  • RAC-KTQ63DPVVB37 — ADR-003 Structured Outputs First
  • RAC-KTQ63DPWHA4B — ADR-004 Artifact Model
  • RAC-KTQ63DPX18Q7 — ADR-005 CLI First
  • RAC-KTQ63DPY3ABD — ADR-006 Ingestion Over Rewrite
  • RAC-KTQ63DPYKJF4 — ADR-007 JSON Contract Stability
  • RAC-KTQ63DPZ42BG — ADR-008 Agent Ready Architecture
  • RAC-KTQ63DQ2AEJZ — ADR-010: Documents Are Not Artifacts
  • RAC-KTQ63DQ3A5FQ — ADR-011: File-First Pipelines
  • RAC-KTQ63DQZ2VSV — ADR-017 — RAC Manages Knowledge, Not Work
  • RAC-KTQ63DR3G4YG — ADR-018: RAC Directory as the Canonical Knowledge Root
  • RAC-KTQ63DR66AMP — ADR-019: Asset References
  • RAC-KTQ63DRA31YC — ADR-020: Requirements as Long-Lived Product Capabilities
  • RAC-KTQ63DRET9QV — ADR-021: Templates as Artifact Creation Contracts
  • RAC-KTQ63DRPK57V — ADR-023: Clean-Break Internal Refactors
  • RAC-KTQ63DRWTN1T — ADR-024: RAC Is Not a Content Store
  • RAC-KTQ63DSC8SZW — ADR-027: CI Test Topology — Merge-Gated, Per-Service Batteries (Process)
  • RAC-KTQ63DSEK9YG — ADR-028 Explorer Delivery Surface
  • RAC-KTW0M8104880 — ADR-029: Guide Delivery Surface (Architecture)
  • RAC-KTW0M8184YYT — ADR-030: Guide Tools-Only Surface (Product)
  • RAC-KTW0M81B0GBB — ADR-031: Guide In-Process Core Consumption (Architecture)
  • RAC-KTW0M81E7TRA — ADR-032: Guide Stateless Reads (Technical)
  • RAC-KTW0M81HX5C6 — ADR-033: Guide Response Budget (Technical)
  • RAC-KTW0M81MVJ7D — ADR-034: Guide Agent Reasoning Boundary (Product)
  • RAC-KTXSTGNKHHVX — ADR-036: Lore Product Identity (Product)
  • RAC-KTXTAF6ZKDK8 — ADR-037: Token-Boundary Search Matching (Technical)
  • RAC-KTXTAG63E89H — ADR-038: Body-Text Search Tier (Technical)
  • RAC-KTY0D0DFTCJA — ADR-039: Lore Server Identity (Product)
  • RAC-KTY25D945HYK — ADR-040: Guide Local Telemetry (Product)
  • RAC-KTYPAB0HWFJD — ADR-041: Anonymous Usage Ping (Product)
  • RAC-KTYXDTB4299E — ADR-042: Documentation Site Hosting (Product)
  • RAC-KTYZVKZQWD98 — ADR-043: Watchkeeper Revision Materialization (Technical)
  • RAC-KV2E5A5E1F1H — ADR-044: Onboarding Scaffold Writes One Starter Artifact (Product)
  • RAC-KV2E5B1122YN — ADR-045: Recency Is Derived From Git, Not Stored In Frontmatter (Technical)
  • RAC-KV2E9HYHVV8G — ADR-046: CLI Usage Telemetry (Product)
  • RAC-KV2J0GYNCAJF — ADR-047: Agent Operating-Guidance Documents Are Prompt Artifacts (Process)
  • RAC-KV4ZAGWPAA6X — ADR-059: Reuse a Single Markdown Parser Instance (Architecture)
  • RAC-KV4ZAHVNGH2J — ADR-060: Share Structural Validation Across Per-Type Validators (Architecture)
  • RAC-KV5112MVD0AM — ADR-061: Roadmaps Carry an "Achieved" Terminal Lifecycle Status (Architecture)
  • RAC-KV5DJYE5FGH0 — ADR-062: The Python SDK's Public Surface Is rac.__all__ (Architecture)
  • RAC-KV68XJGEXBNB — ADR-064: Multi-Repo Extraction Strategy for the itsthelore Organisation (Architecture)
  • RAC-KV6ADYFGC3H4 — ADR-063: Non-Python Clients Are Thin Clients Over the Contract (Architecture)
  • RAC-KV6KFBDZ4D23 — ADR-065: Artifact Content Is Untrusted Input; the Trust Boundary Is Human PR Review (Architecture)
  • RAC-KV6KFCC8MHTM — ADR-066: Grounding Eval Scoring Is Deterministic — No Embeddings, No LLM Judge (Technical)
  • RAC-KV80WX94GY8A — ADR-067: Agent Integration is Context-Supply and Post-Edit Enforcement, Not Pre-Edit Interception (Architecture)
  • RAC-KV8WY8XAJ55S — ADR-070: Prompt-Complexity Routing Boundary (Architecture)
  • RAC-KV8YZQC7G7NW — ADR-069: Wayfinder — Prompt-Complexity Routing as a Separate Product (Product)
  • RAC-KVA44MVMDXXX — ADR-068: Extension, SDK, and Brand Architecture (Architecture)
  • RAC-KVJK92SM2A1R — ADR-072: Document Ingestion Parser Is markitdown (Architecture)
  • RAC-KVJY1KJEWZ87 — ADR-073: Backend Connectors Are Export-Contract Consumers, Not Per-Provider Repos (Architecture)
  • RAC-KVK19NPWFYC9 — ADR-074: The Graph Export Surfaces Typed Relationship Edges (Technical)
  • RAC-KVNM01QPBPXB — ADR-075: The Pre-Merge Check Tier Is a Required Merge Gate on main (Process)
  • RAC-KVSQ2A0BB9XF — ADR-079: Note-Tool Exports Are Ingested by Normalisation, Not markitdown (Architecture)
  • RAC-KVSTYDARXKYW — ADR-080: The Single Source of Truth Is Git, Not a Database (Architecture)
  • RAC-KVTS86ZGVJV7 — ADR-077: The Two-Gate Capture Write Model (Architecture)
  • RAC-KVTSPK4CJHWK — ADR-083: Third-Party Artifact Types via Pinned Spec Bundles (Schema Composition Over the Code-Defined Core) (Architecture)
  • RAC-KW2YW6XK593X — ADR-084: Read-Access Audit Recorder (Product)
  • RAC-KW47GFBHK31W — ADR-086: Air-Gap Posture and Enterprise Telemetry Hard-Lock (Product)
  • RAC-KW47GGS85CKG — ADR-087: External-Reference Relationships (Jira and Beyond) (Technical)
  • RAC-KW47GJ749PKC — ADR-088: Enterprise Profile Scaffold (rac init --profile) (Product)
  • RAC-KW47GN4SB403 — ADR-090: Enterprise Integration Surfaces and Boundaries (Architecture)
  • RAC-KW5A03ZHXN9F — ADR-092: Repository Topology — One Repo Per Concern, rac-* Naming (Architecture)
  • RAC-KW6HY8W1CBK6 — ADR-093: Roadmap Intent Lives in the Corpus; Execution Is Tracked in GitHub Issues (Process)
  • RAC-KW6NQ4Y814N7 — ADR-094: Roadmaps Are Identified by Stable Codename, Not a vX.Y.Z Scope-Fence (Process)
  • RAC-KW8ZTVEQJ366 — ADR-095: rac-connectors Repackage, PyPI Rename, and Outbound Field Rename (Architecture)
  • RAC-KWBJJRZR84PT — ADR-096: External-Target Verification Edge (verified_by) (Technical)
  • RAC-KWFVA38YT2C0 — ADR-097: Benchmark Family Contract (Technical)
  • RAC-KWMW45KXHZJP — ADR-098: Shared HTTP MCP Serving (Architecture)
  • RAC-KWMZ3MR9DZ09 — ADR-099: Derived-Index Cache (Technical)
  • RAC-KWS4Y9KCTD90 — ADR-103: Unified Derived Read-Model (Architecture)
  • RAC-KWS7QCT10Q5A — ADR-104: Persistent Memory-Mapped Index Store (Architecture)
  • RAC-KWSDFYW7PCW6 — ADR-105: Event-Sourced Serving Freshness (Architecture)
  • RAC-KWSH9J2S7QB1 — ADR-106: Incremental Directory Validation (Architecture)
  • RAC-KWSJZJ30EN1J — ADR-107: Parallel Cold Build of the Derived Index (Architecture)
  • RAC-KWV0HZ2SSGE2 — ADR-101: Org-Wide Documentation Site Amends Docs Hosting and Repository Topology (Product)
  • RAC-KWV63BR8AW71 — ADR-102: Org-Site Brand Direction — Light, Restrained, Site Rhymes With Product UI (Product)
  • RAC-KWY0SHYCVY2D — ADR-108: Term-Range-Partitioned Parallel Merge for the Cold Build (Architecture)
  • RAC-KWY7886GSEE5 — ADR-109: Tag Search Tier and Tag Facet (Technical)
  • RAC-KX04DH293JG8 — ADR-111: Revert to SemVer Release Versioning (Process)
  • RAC-KX2WTHEMDEY0 — ADR-112: Cache On by Default, Stat-Proxy Freshness as the Floor (Technical)
  • RAC-KX8GEA45HRBM — ADR-113: Capture Writes Arrive Through a Sibling Surface, Not Guide (Architecture)
  • RAC-KX9H2M7Q4V8C — ADR-122: OKF v0.2 Is a Truthful Derived Carrier (Architecture)
  • RAC-KXBD3T7Q9M2N — ADR-123: Deterministic Decision-to-Code Enforcement (Architecture)
  • RAC-KXE0M2QBF2MP — ADR-114: Native Index Workspace Dependencies — memmap2 In, inotify Deferred (Technical)
  • RAC-KXFK11FQDN1Y — ADR-115: The Shared Artifact-Spec Registry Both Engines Read (ADR-063 Guard 1) (Architecture)
  • RAC-KXGVR299XY5E — ADR-116: The Native Rust Engine Is a Sanctioned Second Implementation Under Lockstep Guards (Architecture)
  • RAC-KYVTHFQD44BP — ADR-124: Publish the Native MCP Server Through OCI and the Official Registry (Architecture)
  • RAC-KYYC7HBFMRBA — ADR-126: Package the Native MCP Server for Docker's MCP Catalog (Architecture)
  • RAC-KZ0F0RG3N5XT — ADR-149: Git Repository Truth Is Forge-Agnostic (Architecture)
  • RAC-KZKMJ8WSMFA1 — ADR-134: Declare and Verify an Offline Materialised Parent (Architecture)
  • RAC-KZKMJ92ABVJG — ADR-135: Use corpus.source as the Global Corpus Identity (Architecture)
  • RAC-KZKMJ9DGR69Z — ADR-137: Require Decision-Backed Explicit Federation Overrides (Architecture)
  • RAC-KZKMJ9K3AFB2 — ADR-138: Build Federation Through One Source-Aware Read Model (Architecture)
  • RAC-KZKMJ9RP0KNV — ADR-139: Rank Federated Retrieval Without Source Preference (Technical)
  • RAC-KZKMJ9YA8BRG — ADR-140: Apply Inherited Decisions to Child Code (Product)
  • RAC-KZKMJA3YK5Y1 — ADR-141: Add Bounded Federation Provenance to the Existing MCP Surface (Product)
  • RAC-KZKMJA9JVF6J — ADR-142: Export the Inherited Layer by Default (Product)
  • RAC-KZN54DB1VNPB — ADR-144: Compose Federation as a Bounded Acyclic Source Graph (Architecture)
  • RAC-KZN54DB2M7FZ — ADR-145: Declare Multiple Offline Parents Through a Versioned Federation Manifest (Architecture)
  • RAC-KZN54DB3V0ZC — ADR-146: Resolve Federated Artifacts by Global Source Identity Without Precedence (Architecture)
  • RAC-KZN54DB4QY0R — ADR-147: Permit Decision-Backed Override Chains Across the Federation Graph (Architecture)
  • RAC-KZN54DB55X9R — ADR-148: Key Serving State to the Entire Federated Closure (Technical)
  • RAC-M32XGEVNX50E — ADR-150: Inherit Parent Spec Bundles Across the Federation Graph (Architecture)
  • RAC-M3C0K38CZYQ1 — ADR-151: Risk Is a Built-In Knowledge Family (Architecture)
  • RAC-MCP20260728A — ADR-121: Dual-Era MCP Protocol Compatibility (Architecture)
  • RAC-P55FRE5HNE55 — ADR-118: Native Event Freshness Acceleration
  • RAC-P61BA5EDE7A0 — ADR-119: Base-Plus-Delta Serving Generations
  • RAC-PYRE71RECER7 — ADR-120: Rust CI Uses Contract Fixtures and Live-Corpus Invariants

Read the full file on GitHub · 149 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. 1mo ago Changed · +1 lines · +27 tokens per session a614728f0836
  2. 1mo ago Changed · +2 lines · +77 tokens per session a3c2631218dc
  3. 2mo ago First seen · 146 lines · 3,791 tokens per session scan A bbf4258c89b4

Subscribe to this mod's changes

core CLAUDE.md is an instructions file published in the GitHub repository asdecided/core (300 stars, last pushed yesterday), licensed Apache-2.0. It adds 3,895 tokens to every session, about $0.0156 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 94% identical to core copilot-instructions.md, differing in 25 lines, and is treated as a copy.

Related

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).

microsoft/vscode · 9,715 tokens

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.

vercel/next.js · 7,461 tokens

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).

microsoft/vscode · 5,262 tokens

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.

langchain-ai/langchain · 4,680 tokens

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).

microsoft/vscode · 2,750 tokens

deepseek-harness AGENTS.md

AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.

deepseek-ai/deepseek-harness · 4,261 tokens