Skip to main content

AI Agent Config Standards

How do independent AI agents coordinate on the same codebase without vendor lock-in?

Agentic AI Foundation

The Agentic AI Foundation (AAIF) launched December 2025 under the Linux Foundation. Three anchor standards, donated by competitors who agree coordination beats fragmentation.

StandardDonated ByGovernsMaturity
AGENTS.mdOpenAI / SourcegraphProject instructionsAdopted: 60,000+ repos, 20+ tools
Agent SkillsAnthropicModular capabilitiesEarly: Codex native, Copilot adopting
MCPAnthropicTool connectivityAdopted: Claude, Cursor, VS Code, Gemini

Platinum members: Anthropic, OpenAI, Block, AWS, Bloomberg, Cloudflare, Google, Microsoft.

Complementary: Agent Client Protocol (ACP) by JetBrains and Zed standardizes how agents talk to IDEs — the transport layer, not the config file layer.

Three Layers

AGENTS.md              WHAT to do in this repo    (instructions)
.agents/skills/ HOW to do specific tasks (capabilities)
MCP config WHICH external tools (services)

These compose. AGENTS.md says "follow our content standards." .agents/skills/content-flow/SKILL.md provides the procedure. MCP connects the validator tool.

AGENTS.md Spec

The project instruction file. Tells every agent what matters in this repo.

PropertyDetail
FileAGENTS.md at project root and any subdirectory
OverrideAGENTS.override.md takes precedence (same locations)
Global~/.codex/AGENTS.md for user-wide defaults
Size limit32 KiB combined (configurable via project_doc_max_bytes)
WalkRoot → cwd, each directory checked, files concatenate

Which tools read AGENTS.md:

ToolReads AGENTS.mdNative FileNotes
Codex CLIYes (primary)AGENTS.mdFirst mover
GitHub CopilotYes.github/copilot-instructions.mdDetects AGENTS.md, CLAUDE.md, GEMINI.md
CursorYes.cursor/rules/*.mdcReads alongside native rules
Gemini CLIYesGEMINI.mdFalls back to AGENTS.md
Claude CodeNoCLAUDE.mdFeature request #6235 (2,777+ upvotes)
WindsurfNot confirmed.windsurf/rules/*.md

The gap: Claude Code is the notable holdout. Workaround: @AGENTS.md import in CLAUDE.md.

Agent Skills Spec

Authored by Anthropic (Barry Zhang, Keith Lazuka, Mahesh Murag). Published at agentskills.io.

Structure

.agents/
skills/
content-flow/
SKILL.md # Required: name + description frontmatter
references/ # Optional: loaded on demand
cookbook/ # Optional: examples and patterns
scripts/ # Optional: executable code
assets/ # Optional: templates, schemas

SKILL.md Frontmatter

---
name: content-flow
description: Write and review /docs pages for Structure, Substance, and Connection
argument-hint: /path/to/file.md
---
FieldRequiredMax LengthPurpose
nameYes64 charsLowercase + hyphens. Discovery key.
descriptionYes1024 charsWhen to activate. The trigger.
argument-hintNoWhat to pass when invoking
licenseNoFor shared/published skills
compatibilityNoWhich agents support this skill
allowed-toolsNoRestrict tool access

Token Budget

Progressive disclosure controls context cost:

StageWhat LoadsBudget
Startupname + description from all skills~100 tokens per skill
ActivationFull SKILL.md body<5,000 tokens recommended
On demandreferences/, cookbook/, scripts/Loaded via Read when needed

Discovery

Agents find skills by scanning .agents/skills/ from cwd upward to repo root. Codex also scans:

  • ~/.codex/skills/ (user-level)
  • ~/.codex/skills/.system/ (system-level, admin-deployed)

Model Context Protocol

MCP standardizes how agents connect to external tools (databases, APIs, file systems). It defines servers, tools, resources, and prompts — but not where configs live.

ToolMCP Config Location
Claude Code.mcp.json (project) or ~/.claude.json (user)
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
JetBrainsIDE-level mcp.json

No .mcp/ directory convention exists. Each IDE owns its config location. MCP standardizes the protocol, not the file layout.

Tool Config Matrix

Complete mapping of where each tool reads configuration.

Config TypeClaude CodeGemini CLICodex CLICursorWindsurfGitHub Copilot
Root fileCLAUDE.mdGEMINI.mdAGENTS.md.cursor/rules/*.mdc.windsurfrules.github/copilot-instructions.md
Nested contextCLAUDE.md per dirGEMINI.md per dirAGENTS.md per dirGlobs onlyTo git rootExperimental
Rules dir.claude/rules/*.md.cursor/rules/*.mdc.windsurf/rules/*.md.github/instructions/*.instructions.md
Glob scopingpaths: frontmatterglobs: frontmatterActivation modesapplyTo: frontmatter
Commands.claude/commands/*.md.gemini/commands/*.toml.agents/skills/.cursor/agents/*.md.windsurf/workflows/*.md.github/prompts/*.prompt.md
Global config~/.claude/CLAUDE.md~/.gemini/GEMINI.md~/.codex/AGENTS.mdUser Rules (UI)global_rules.md (UI)VS Code settings
Auto memoryMEMORY.md (auto)/memory add (manual)Notepads (beta)Cascade Memories
Import syntax@path@file.md#file: / links
MCP.mcp.jsongemini-extension.jsonSettingsAgent frontmatter

Standard Maturity

How mature is each standard? When to adopt vs wait.

StandardMaturitySignalOur Position
AGENTS.mdAdopted60K+ repos, AAIF governance, 20+ toolsAdopt: maintain AGENTS.md at root
Agent Skills (.agents/skills/)Early adopterSpec published, Codex native, Copilot adoptingAdopt: skills in .agents/skills/
MCPAdoptedBroad IDE support, AAIF governanceAdopt: .mcp.json configured
ACP (IDE transport)EmergingJetBrains + Zed onlyWatch: not relevant until IDE diversity matters
.agent/ directory (issue #71)ProposalOpen GitHub issue, mixed feedbackWait: premature, conflicts with .agents/
Unified rules directoryNot proposedEach tool has its own formatWait: symlink from .ai/rules/ to agent-native dirs

Adoption Decision Framework

When a new standard appears, run this test:

GateQuestionAdopt IfWait If
GovernanceWho owns it?Foundation (AAIF, Linux Foundation)Single vendor
AdoptionWho uses it?3+ major tools1 tool only
StabilityDoes the spec change often?Versioned, backwards-compatibleBreaking changes monthly
CostWhat breaks if we adopt?Adapter only, no rewriteFull migration required
ReversibilityCan we undo it?Rename/symlink backData loss or lock-in

Applied to Agent Skills: Governance (AAIF) pass. Adoption (Codex + Copilot) pass. Stability (v1 spec published) pass. Cost (rename .ai/skills/ to .agents/skills/, update references) low. Reversibility (rename back) trivial. Decision: adopt.

Our Implementation

See Config Architecture for full documentation of how we implement these standards:

  • AGENTS.md — self-contained project instructions for Codex/Copilot
  • .agents/skills/ — 13 skills with standard SKILL.md frontmatter
  • .ai/ — rules, hooks, commands, voices (no standard yet)
  • CLAUDE.md / GEMINI.md — agent-specific routers via @-imports
  • .claude/commands/ — proxy files redirect to .ai/commands/
  • .claude/rules/ — symlinks to .ai/rules/

Context