Add cross-tool AI distribution: npx skills, llms.txt index, Context7 config

.claude/.codebase-info/.map-state.json

@@ -2,7 +2,7 @@
   "tool": "codebase-mapper",
   "version": "2.8.0",
   "mappedAt": "2026-07-18",
-  "gitCommit": "7e85456d9984483b29211bff1dd98820409e9635",
+  "gitCommit": "ffcdca567bd460af080734598762112a9af6857e",
   "documents": [
     "architecture.md",
     "tech-landscape.md",
@@ -18,8 +18,8 @@
   "hashes": {
     "INDEX.md": "17eaa0ba82f159465e126597aac349c6e4429330887b58742636734e15854313",
     "architecture.md": "7c908f047e6b0f08824ccd7c3f674ad47bd547dde234d0bd424a6d2784c0cbb6",
-    "tech-landscape.md": "a9052f4bba3742e79556a49c888424e6d9de424970e9def440c7124f45ec6a41",
-    "directory-structure.md": "ecf7e3357b41a83e7f31e760db41ef0bcb4fdbc6f1d61bbb4daf896f425d6657",
+    "tech-landscape.md": "cad1a11e5e63013a10cf07657d248196d0e3e4e14203eba7b777c1c123628198",
+    "directory-structure.md": "7d04bf29e9005508af316e29ded0c2f082a8dca561a279faecbc682f3bf5db6f",
     "entry-points.md": "6bd4ccae3d975196e03fc2c4e86007eba89f1f6427232fd8adce78ec97d6b432",
     "modules.md": "1238c76664edb1c1667fda23f39507bcab5b760b57a3a3728d7104b2cf8032e7",
     "communication.md": "49741a7455b0f162881e1cfb2d797dc8e3832cc8782796bbcb8a10529f2c5a0b",

.claude/.codebase-info/directory-structure.md

@@ -22,8 +22,9 @@ atomic-agents/                  # repo root (uv workspace)
 ├── atomic-examples/            # 16 runnable example apps (each its own project)
 ├── docs/                       # Sphinx + MyST documentation (api/, guides/, examples/)
 ├── guides/                     # DEV_GUIDE.md and contributor guides
-├── scripts/                    # generate_llms_files.py (llms-*.txt bundles for the docs site)
+├── scripts/                    # generate_llms_files.py (llms.txt index + llms-*.txt bundles)
 ├── pyproject.toml              # package metadata, deps, [tool.black], uv workspace
+├── context7.json               # Context7 indexing config + v2 API rules for AI assistants
 ├── build_and_deploy.ps1        # version bump + uv build/publish
 ├── AGENTS.md                   # the project's own design philosophy (imported by CLAUDE.md)
 └── README.md

.claude/.codebase-info/tech-landscape.md

@@ -39,9 +39,9 @@
     `pytest --cov=atomic_agents` across `atomic-agents`, `atomic-assembler`, `atomic-examples`,
     `atomic-forge`.
   - `docs.yml` — on push to `main`: build Sphinx docs, run
-    `scripts/generate_llms_files.py` (produces `llms-*.txt`), deploy to GitHub Pages.
-    The docs version needs no sync step — `docs/conf.py` reads it from the root
-    `pyproject.toml` via `tomllib`.
+    `scripts/generate_llms_files.py` (spec-compliant `llms.txt` index + `llms-*.txt`
+    bundles, copied to the site root), deploy to GitHub Pages. The docs version needs
+    no sync step — `docs/conf.py` reads it from the root `pyproject.toml` via `tomllib`.
 - **Release:** `build_and_deploy.ps1 <major|minor|patch> [--dry]` bumps the version in
   `pyproject.toml`, then `uv sync` → `uv build` → `uv publish` (requires a `PYPI_TOKEN`).
 - **Distribution:** PyPI package `atomic-agents`; the `atomic` console script ships with it.

.claude/skills/release/SKILL.md

@@ -2,6 +2,8 @@
 name: release
 description: Release a new version of atomic-agents to PyPI and GitHub. Use when the user asks to "release", "publish", "deploy", or "bump version" for atomic-agents.
 allowed-tools: Read, Bash, Grep, Glob
+metadata:
+  internal: true
 ---
 
 # Release Process for atomic-agents

.github/workflows/docs.yml

@@ -57,11 +57,16 @@ jobs:
           cd docs
           uv run make html
 
+      - name: Place llms files at site root
+        run: |
+          cp docs/llms.txt docs/llms-*.txt docs/_build/html/
+
       - name: Verify llms files in build
         run: |
-          echo "Checking for llms files in build directory:"
+          echo "Checking for llms files at site root:"
+          ls -lh docs/_build/html/llms*.txt
+          echo "Checking for llms files in _static (download links):"
           ls -lh docs/_build/html/_static/llms-*.txt || echo "Files not in _static"
-          ls -lh docs/_build/html/_downloads/*/llms-*.txt || echo "Files not in _downloads"
 
       - name: Upload artifact
         uses: actions/upload-pages-artifact@v3

README.md

@@ -44,6 +44,9 @@ Built on [Instructor](https://github.com/jxnl/instructor) and [Pydantic](https:/
   - [Examples \& Documentation](#examples--documentation)
     - [Quickstart Examples](#quickstart-examples)
     - [Complete Examples](#complete-examples)
+  - [AI-Assisted Development](#ai-assisted-development)
+    - [Agent skills (Claude Code, Cursor, Copilot, Codex, Windsurf, Gemini CLI, ...)](#agent-skills-claude-code-cursor-copilot-codex-windsurf-gemini-cli-)
+    - [Docs for LLMs](#docs-for-llms)
   - [🚀 Version 2.0 Released!](#-version-20-released)
     - [Key Changes in v2.0:](#key-changes-in-v20)
     - [⚠️ Upgrading from v1.x](#️-upgrading-from-v1x)
@@ -293,6 +296,35 @@ In addition to the quickstart examples, we have more complex examples demonstrat
 
 For a complete list of examples, see the [examples directory](/atomic-examples/).
 
+## AI-Assisted Development
+
+Building with an AI coding assistant? Atomic Agents ships first-class support so your assistant knows the framework's current API and conventions instead of guessing.
+
+### Agent skills (Claude Code, Cursor, Copilot, Codex, Windsurf, Gemini CLI, ...)
+
+Six [agent skills](https://agents.md/) cover the framework: an auto-triggering `framework` guide with progressive-disclosure references, four creation workflows (`create-atomic-schema`, `create-atomic-agent`, `create-atomic-tool`, `create-atomic-context-provider`), and a `new-app` project scaffolder.
+
+**Claude Code** (also gets the `atomic-explorer` and `atomic-reviewer` subagents):
+
+```
+/plugin marketplace add eigenwise/atomic-agents
+/plugin install atomic-agents@eigenwise
+```
+
+**Any other skills-compatible assistant:**
+
+```bash
+npx skills add eigenwise/atomic-agents
+```
+
+See [claude-plugin/atomic-agents](./claude-plugin/atomic-agents/README.md) for details.
+
+### Docs for LLMs
+
+The [documentation site](https://eigenwise.github.io/atomic-agents/) publishes LLM-ready bundles: [llms-full.txt](https://eigenwise.github.io/atomic-agents/llms-full.txt) (docs + source + examples), plus separate [docs](https://eigenwise.github.io/atomic-agents/llms-docs.txt), [source](https://eigenwise.github.io/atomic-agents/llms-source.txt), and [examples](https://eigenwise.github.io/atomic-agents/llms-examples.txt) bundles. Point your assistant at one of these when it needs the full picture.
+
+Atomic Agents is also indexed on [Context7](https://context7.com/eigenwise/atomic-agents) for assistants with the Context7 MCP server.
+
 ## Atomic Forge & CLI
 
 Atomic Forge is a collection of tools that can be used with Atomic Agents to extend its functionality. Current tools include:

claude-plugin/atomic-agents/README.md

@@ -54,6 +54,16 @@ To require this plugin for everyone who opens a given repo, add it to the projec
 
 Claude Code prompts each teammate to install on first trust of the folder.
 
+### Other coding assistants (Cursor, Copilot, Codex, Windsurf, Gemini CLI, ...)
+
+The skills follow the cross-tool [agent skills](https://agents.md/) format, so any skills-compatible assistant can install them without the Claude Code plugin system:
+
+```bash
+npx skills add eigenwise/atomic-agents
+```
+
+This installs the six skills (framework guide, four create-* workflows, new-app scaffolder). The `atomic-explorer` and `atomic-reviewer` subagents are Claude Code plugin features and don't travel this path.
+
 ### Local dev
 
 ```bash

context7.json

@@ -0,0 +1,16 @@
+{
+  "$schema": "https://context7.com/schema/context7.json",
+  "projectTitle": "Atomic Agents",
+  "description": "Lightweight, modular Python framework for building agentic AI applications with typed Pydantic schemas and Instructor",
+  "branch": "main",
+  "excludeFolders": [".claude", ".github", "docs/_build", "**/__pycache__"],
+  "excludeFiles": ["CHANGELOG.md"],
+  "rules": [
+    "Import from the top-level package: from atomic_agents import AtomicAgent, AgentConfig, BaseIOSchema, BaseTool. Legacy paths like atomic_agents.lib.* were removed in v2.",
+    "Construct agents as AtomicAgent[InputSchema, OutputSchema](config=AgentConfig(...)). The type parameters carry runtime information, so write them explicitly.",
+    "Always wrap the provider SDK client with Instructor (instructor.from_openai, instructor.from_anthropic, ...) before passing it as AgentConfig.client.",
+    "Every BaseIOSchema subclass requires a non-empty docstring; the docstring and Field descriptions flow into the LLM prompt.",
+    "Provider parameters (temperature, max_tokens, reasoning_effort) go in AgentConfig.model_api_parameters, not on the agent.",
+    "For Gemini set assistant_role='model' and match AgentConfig.mode to the Instructor factory mode; Anthropic requires max_tokens in model_api_parameters."
+  ]
+}

scripts/generate_llms_files.py

@@ -22,11 +22,14 @@
 SINGLEHTML_DIR = DOCS_DIR / "_build" / "singlehtml"
 
 # Output files
+OUTPUT_INDEX = DOCS_DIR / "llms.txt"
 OUTPUT_DOCS = DOCS_DIR / "llms-docs.txt"
 OUTPUT_SOURCE = DOCS_DIR / "llms-source.txt"
 OUTPUT_EXAMPLES = DOCS_DIR / "llms-examples.txt"
 OUTPUT_FULL = DOCS_DIR / "llms-full.txt"
 
+SITE_URL = "https://eigenwise.github.io/atomic-agents"
+
 # File extensions to include in source code
 SOURCE_CODE_EXTENSIONS = {'.py', '.md', '.txt', '.yml', '.yaml', '.toml', '.cfg', '.ini', '.json'}
 
@@ -240,6 +243,46 @@ def generate_full_content() -> str:
 
     return content
 
+def generate_index() -> str:
+    """Generate the llms.txt index per the llmstxt.org spec: a small Markdown
+    navigation file at the site root pointing to the full bundles and key pages."""
+    return f"""# Atomic Agents
+
+> Atomic Agents is a lightweight, modular Python framework for building agentic AI applications
+> as composable, schema-driven building blocks. Built on Instructor and Pydantic: every agent is
+> a typed transformer from an input schema to an output schema, with full developer control and
+> no hidden abstractions. Provider-agnostic (OpenAI, Anthropic, Groq, Gemini, Ollama, and more).
+
+Key API facts: import from the top-level `atomic_agents` package (`AtomicAgent`, `AgentConfig`,
+`BaseIOSchema`, `BaseTool`) and from `atomic_agents.context` (`ChatHistory`,
+`SystemPromptGenerator`, `BaseDynamicContextProvider`). Agents are constructed as
+`AtomicAgent[InputSchema, OutputSchema](config=AgentConfig(...))` with an Instructor-wrapped
+client. Every `BaseIOSchema` subclass requires a docstring. Python 3.12+.
+
+## Documentation
+
+- [Quickstart guide]({SITE_URL}/guides/quickstart.html): first agent, providers, streaming
+- [Tools guide]({SITE_URL}/guides/tools.html): BaseTool, Atomic Forge, MCP interop
+- [Orchestration guide]({SITE_URL}/guides/orchestration.html): multi-agent patterns
+- [Memory guide]({SITE_URL}/guides/memory.html): ChatHistory, persistence, custom backends
+- [Hooks guide]({SITE_URL}/guides/hooks.html): telemetry, retries, error handling
+- [API reference]({SITE_URL}/api/index.html): full API documentation
+
+## Bundles
+
+- [llms-full.txt]({SITE_URL}/llms-full.txt): complete documentation, framework source code, and all examples in one file
+- [llms-docs.txt]({SITE_URL}/llms-docs.txt): documentation only
+- [llms-source.txt]({SITE_URL}/llms-source.txt): framework source code only
+- [llms-examples.txt]({SITE_URL}/llms-examples.txt): runnable example projects only
+
+## Optional
+
+- [GitHub repository](https://github.com/eigenwise/atomic-agents): source, issues, discussions
+- [Agent skills](https://github.com/eigenwise/atomic-agents/tree/main/claude-plugin/atomic-agents): installable skills for coding assistants (`npx skills add eigenwise/atomic-agents`)
+- [PyPI package](https://pypi.org/project/atomic-agents/): `pip install atomic-agents`
+"""
+
+
 def write_to_file_and_copy(output_file: Path, content: str):
     """Write content to file and copy to HTML build directory if it exists."""
     with open(output_file, "w", encoding="utf-8") as f:
@@ -277,6 +320,10 @@ def main():
     print("\nGenerating llms-full.txt...")
     full_content = generate_full_content()
     write_to_file_and_copy(OUTPUT_FULL, full_content)
+
+    # Generate the spec-compliant index
+    print("\nGenerating llms.txt index...")
+    write_to_file_and_copy(OUTPUT_INDEX, generate_index())
 
     print("\n" + "=" * 40)
     print("Successfully generated all llms.txt files!")