Remove Agent and AI APIs (#165)

## Summary
- remove the Agent/OpenAI API surface (`Codebase.ai`, `set_ai_key`,
`max_ai_requests`, OpenAI secrets/dependency, and related errors/tests)
- delete AI-oriented extensions and examples: semantic/vector index, MCP
servers, generated system prompt resources, SWE-bench agent harness,
attribution/AI impact, RAG/document/training-data examples
- remove Agent run paths/schemas from `docs/api-reference/openapi3.json`
and clean live docs/CLI copy so codemod scaffolding is deterministic
only
- refresh root and examples lockfiles after removing
OpenAI/MCP/datasets/lox dependency stacks

## Testing
- `uv run ruff format $(git diff --name-only --diff-filter=ACM | rg
'\.py$' | tr '\n' ' ')`
- `uv run ruff check $(git diff --name-only --diff-filter=ACM | rg
'\.py$' | tr '\n' ' ')`
- `uv run pytest tests/unit/sdk/python/codebase/test_control_flow.py
tests/unit/runner/sandbox/test_executor.py
tests/integration/codegen/sdk/code_generation/test_api_doc_generation.py`
- `uv run pytest tests/unit/cli/commands/parse/test_parse.py
tests/unit/cli/commands/diagnose/test_diagnose.py
tests/unit/cli/commands/run/test_run.py tests/unit/gscli/test_cli.py`
- `jq empty docs/api-reference/openapi3.json`
- `uv run python -c "import graph_sitter; from graph_sitter import
Codebase; import graph_sitter.extensions; import graph_sitter.cli.cli;
print('imports ok')"`
- `PATH="$HOME/.nvm/versions/node/v22.19.0/bin:$PATH" npm --prefix site
ci`
- `PATH="$HOME/.nvm/versions/node/v22.19.0/bin:$PATH" npm --prefix site
run build`

## Not run
- `uv run pytest tests/integration/codemod/test_verified_codemods.py
--collect-only -q` is blocked by missing fixture data:
`tests/integration/verified_codemods/codemod_data/Y2Y4MGNkOG.json`.

---------

Co-authored-by: jayhack <2548876+jayhack@users.noreply.github.com>

Author: jayhack

docs/api-reference/core/Codebase.mdx

@@ -87,41 +87,6 @@ import {Attribute} from '/snippets/Attribute.mdx';
 
 ## Methods
 <HorizontalDivider />
-### <span className="text-primary">ai</span>
-Generates a response from the AI based on the provided prompt, target, and context.
-<GithubLinkNote link="https://github.com/codegen-sh/graph-sitter/blob/develop/src/graph_sitter/core/codebase.py#L1196-L1283" />
-
-<ParameterWrapper>
-<Parameter
-    name="prompt"
-    type={ <code className="text-sm bg-gray-100 px-2 py-0.5 rounded">str</code> }
-    description="The text prompt to send to the AI."
-    defaultValue=""
-/>
-<Parameter
-    name="target"
-    type={ <><a href="/api-reference/core/Editable" style={ {fontWeight: "inherit", fontSize: "inherit"} }>Editable</a> <span>| None</span></> }
-    description="An optional editable object (like a function, class, etc.) that provides the main focus for the AI's response."
-    defaultValue="None"
-/>
-<Parameter
-    name="context"
-    type={ <><a href="/api-reference/core/Editable" style={ {fontWeight: "inherit", fontSize: "inherit"} }>Editable</a> <span>| list[</span> <a href="/api-reference/core/Editable" style={ {fontWeight: "inherit", fontSize: "inherit"} }>Editable</a> <span>] | dict[str,</span> <a href="/api-reference/core/Editable" style={ {fontWeight: "inherit", fontSize: "inherit"} }>Editable</a> <span>| list[</span> <a href="/api-reference/core/Editable" style={ {fontWeight: "inherit", fontSize: "inherit"} }>Editable</a> <span>]] | None</span></> }
-    description="Additional context to help inform the AI's response."
-    defaultValue="None"
-/>
-<Parameter
-    name="model"
-    type={ <code className="text-sm bg-gray-100 px-2 py-0.5 rounded">str</code> }
-    description="The AI model to use for generating the response. Defaults to &quot;gpt-4o&quot;."
-    defaultValue="&quot;gpt-4o&quot;"
-/>
-</ParameterWrapper>
-
-
-<Return return_type={ <code className="text-sm bg-gray-100 px-2 py-0.5 rounded">str</code> } description="The generated response from the AI."/>
-
-
 ### <span className="text-primary">checkout</span>
 Checks out a git branch or commit and syncs the codebase graph to the new state.
 <GithubLinkNote link="https://github.com/codegen-sh/graph-sitter/blob/develop/src/graph_sitter/core/codebase.py#L865-L909" />
@@ -622,14 +587,6 @@ Resets the codebase by
 <Return return_type={ <code className="text-sm bg-gray-100 px-2 py-0.5 rounded">None</code> } description=""/>
 
 
-### <span className="text-primary">set_ai_key</span>
-Sets the OpenAI key for the current Codebase instance.
-<GithubLinkNote link="https://github.com/codegen-sh/graph-sitter/blob/develop/src/graph_sitter/core/codebase.py#L1285-L1291" />
-
-
-<Return return_type={ <code className="text-sm bg-gray-100 px-2 py-0.5 rounded">None</code> } description=""/>
-
-
 ### <span className="text-primary">set_session_options</span>
 Sets the session options for the current codebase.
 <GithubLinkNote link="https://github.com/codegen-sh/graph-sitter/blob/develop/src/graph_sitter/core/codebase.py#L1310-L1328" />
@@ -647,12 +604,6 @@ Sets the session options for the current codebase.
     description="The maximum duration in seconds for a session"
     defaultValue=""
 />
-<Parameter
-    name="max_ai_requests"
-    type={ <code className="text-sm bg-gray-100 px-2 py-0.5 rounded">int, optional</code> }
-    description="The maximum number of AI requests"
-    defaultValue=""
-/>
 </ParameterWrapper>
 
 
@@ -699,4 +650,3 @@ Visualizes a NetworkX graph or Plotly figure.
 <Return return_type={ <code className="text-sm bg-gray-100 px-2 py-0.5 rounded">None</code> } description=""/>
 
 
-

docs/api-reference/openapi3.json

@@ -165,160 +165,6 @@
 				}
 			}
 		},
-		"/v1/organizations/{org_id}/agent/run": {
-			"post": {
-				"tags": ["agents", "agents", "agents"],
-				"summary": "Create Agent Run",
-				"description": "Create a new agent run.\n\nCreates and initiates a long-running agent process based on the provided prompt.\nThe process will complete asynchronously, and the response contains the agent run ID\nwhich can be used to check the status later. The requesting user must be a member\nof the specified organization.\n\nRate limit: 10 requests per minute.",
-				"operationId": "create_agent_run_v1_organizations__org_id__agent_run_post",
-				"parameters": [
-					{
-						"name": "org_id",
-						"in": "path",
-						"required": true,
-						"schema": { "type": "integer", "title": "Org Id" }
-					},
-					{
-						"name": "authorization",
-						"in": "header",
-						"required": false,
-						"schema": { "title": "Authorization" }
-					}
-				],
-				"requestBody": {
-					"required": true,
-					"content": {
-						"application/json": {
-							"schema": { "$ref": "#/components/schemas/CreateAgentRunInput" }
-						}
-					}
-				},
-				"responses": {
-					"200": {
-						"description": "Successful Response",
-						"content": {
-							"application/json": {
-								"schema": { "$ref": "#/components/schemas/AgentRunResponse" }
-							}
-						}
-					},
-					"429": {
-						"content": {
-							"application/json": {
-								"schema": {
-									"$ref": "#/components/schemas/APIRateLimitErrorResponse"
-								}
-							}
-						},
-						"description": "Too Many Requests"
-					},
-					"402": {
-						"content": {
-							"application/json": {
-								"schema": {
-									"$ref": "#/components/schemas/AgentRunLimitReachedErrorResponse"
-								}
-							}
-						},
-						"description": "Payment Required"
-					},
-					"403": {
-						"content": {
-							"application/json": {
-								"schema": {
-									"$ref": "#/components/schemas/PermissionsErrorResponse"
-								}
-							}
-						},
-						"description": "Forbidden"
-					},
-					"422": {
-						"description": "Validation Error",
-						"content": {
-							"application/json": {
-								"schema": { "$ref": "#/components/schemas/HTTPValidationError" }
-							}
-						}
-					}
-				}
-			}
-		},
-		"/v1/organizations/{org_id}/agent/run/{agent_run_id}": {
-			"get": {
-				"tags": ["agents", "agents", "agents"],
-				"summary": "Get Agent Run",
-				"description": "Retrieve the status and result of an agent run.\n\nReturns the current status, progress, and any available results for the specified agent run.\nThe agent run must belong to the specified organization. If the agent run is still in progress,\nthis endpoint can be polled to check for completion.\n\nRate limit: 60 requests per 30 seconds.",
-				"operationId": "get_agent_run_v1_organizations__org_id__agent_run__agent_run_id__get",
-				"parameters": [
-					{
-						"name": "agent_run_id",
-						"in": "path",
-						"required": true,
-						"schema": { "type": "integer", "title": "Agent Run Id" }
-					},
-					{
-						"name": "org_id",
-						"in": "path",
-						"required": true,
-						"schema": { "type": "integer", "title": "Org Id" }
-					},
-					{
-						"name": "authorization",
-						"in": "header",
-						"required": false,
-						"schema": { "title": "Authorization" }
-					}
-				],
-				"responses": {
-					"200": {
-						"description": "Successful Response",
-						"content": {
-							"application/json": {
-								"schema": { "$ref": "#/components/schemas/AgentRunResponse" }
-							}
-						}
-					},
-					"429": {
-						"content": {
-							"application/json": {
-								"schema": {
-									"$ref": "#/components/schemas/APIRateLimitErrorResponse"
-								}
-							}
-						},
-						"description": "Too Many Requests"
-					},
-					"403": {
-						"content": {
-							"application/json": {
-								"schema": {
-									"$ref": "#/components/schemas/PermissionsErrorResponse"
-								}
-							}
-						},
-						"description": "Forbidden"
-					},
-					"404": {
-						"content": {
-							"application/json": {
-								"schema": {
-									"$ref": "#/components/schemas/AgentRunNotFoundErrorResponse"
-								}
-							}
-						},
-						"description": "Not Found"
-					},
-					"422": {
-						"description": "Validation Error",
-						"content": {
-							"application/json": {
-								"schema": { "$ref": "#/components/schemas/HTTPValidationError" }
-							}
-						}
-					}
-				}
-			}
-		},
 		"/v1/organizations": {
 			"get": {
 				"tags": ["organizations", "organizations", "organizations"],
@@ -407,70 +253,6 @@
 				"type": "object",
 				"title": "APIRateLimitErrorResponse"
 			},
-			"AgentRunLimitReachedErrorResponse": {
-				"properties": {
-					"message": {
-						"type": "string",
-						"title": "Message",
-						"default": "Alloted agent runs for the current billing plan have been reached. Please upgrade your plan to continue."
-					},
-					"status_code": {
-						"type": "integer",
-						"title": "Status Code",
-						"default": 402
-					}
-				},
-				"type": "object",
-				"title": "AgentRunLimitReachedErrorResponse"
-			},
-			"AgentRunNotFoundErrorResponse": {
-				"properties": {
-					"message": {
-						"type": "string",
-						"title": "Message",
-						"default": "Agent run not found."
-					},
-					"status_code": {
-						"type": "integer",
-						"title": "Status Code",
-						"default": 404
-					}
-				},
-				"type": "object",
-				"title": "AgentRunNotFoundErrorResponse"
-			},
-			"AgentRunResponse": {
-				"properties": {
-					"id": { "type": "integer", "title": "Id" },
-					"organization_id": { "type": "integer", "title": "Organization Id" },
-					"status": {
-						"anyOf": [{ "type": "string" }, { "type": "null" }],
-						"title": "Status"
-					},
-					"created_at": {
-						"anyOf": [{ "type": "string" }, { "type": "null" }],
-						"title": "Created At"
-					},
-					"web_url": {
-						"anyOf": [{ "type": "string" }, { "type": "null" }],
-						"title": "Web Url"
-					},
-					"result": {
-						"anyOf": [{ "type": "string" }, { "type": "null" }],
-						"title": "Result"
-					}
-				},
-				"type": "object",
-				"required": ["id", "organization_id"],
-				"title": "AgentRunResponse",
-				"description": "Represents an agent run in API responses"
-			},
-			"CreateAgentRunInput": {
-				"properties": { "prompt": { "type": "string", "title": "Prompt" } },
-				"type": "object",
-				"required": ["prompt"],
-				"title": "CreateAgentRunInput"
-			},
 			"HTTPValidationError": {
 				"properties": {
 					"detail": {

docs/building-with-graph-sitter/codegen-with-wsl.mdx

@@ -58,13 +58,9 @@ To use WSL with VSCode:
 
 You can read more about Developing in WSL [here](https://code.visualstudio.com/docs/remote/wsl).
 
-<Tip>
-If you plan on using Codegen's MCP (Model Context Protocol) or LSP (Language Server Protocol) features, running VSCode through WSL is highly recommended.
-</Tip>
-
 ## Troubleshooting
 
-- **I have trouble connecting to MCP or LSP from VSCode**: Windows and WSL run on two different network environments. WSL does come with a compatibility layer called [NAT (Network Address Translation)](https://learn.microsoft.com/en-us/windows/wsl/networking#default-networking-mode-nat) as a translation layer between the two environments, but it can sometimes cause issues. On machines running Windows 11 22H2 and higher, try switching to [Mirrored mode networking](https://learn.microsoft.com/en-us/windows/wsl/networking#mirrored-mode-networking) to see if that fixes the issue.
+- **I have trouble connecting development tools from VSCode**: Windows and WSL run on two different network environments. WSL does come with a compatibility layer called [NAT (Network Address Translation)](https://learn.microsoft.com/en-us/windows/wsl/networking#default-networking-mode-nat) as a translation layer between the two environments, but it can sometimes cause issues. On machines running Windows 11 22H2 and higher, try switching to [Mirrored mode networking](https://learn.microsoft.com/en-us/windows/wsl/networking#mirrored-mode-networking) to see if that fixes the issue.
 - **WSL takes up too much memory after running Codegen**: This is a known issue with WSL. Memory does not get properly released after running a memory-intensive process like Codegen. Try killing the WSL process using `wsl --shutdown` to free up memory. ([GitHub Issue for WSL](https://github.com/microsoft/WSL/issues/4166))
 - **WSL hangs and pins the CPU at 100% when I try to run Codegen**: This could be because of WSL running out of memory. Try killing the WSL process using `wsl --shutdown` and then try running Graph-sitter again. If that doesn't work, try [downloading more RAM](https://downloadmoreram.com/).
 

docs/building-with-graph-sitter/comments-and-docstrings.mdx

@@ -147,7 +147,7 @@ for file in codebase.files:
   indentation and maintains the existing comment style.
 </Note>
 
-## Special APIs and AI Integration
+## Special APIs
 
 ### Google Style Docstrings
 
@@ -160,31 +160,6 @@ func_b = symbol_a.get_method("funcB")
 func_b.docstring.to_google_docstring(func_b)
 ```
 
-### Using AI for Documentation
-
-Graph-sitter integrates with LLMs to help generate and improve documentation. You can use the [Codebase.ai(...)](/api-reference/core/Codebase#ai) method to:
-
-- Generate comprehensive docstrings
-- Update existing documentation
-- Convert between documentation styles
-- Add parameter descriptions
-
-```python
-# Generate a docstring using AI
-function = codebase.get_function("my_function")
-
-new_docstring = codebase.ai(
-    "Generate a comprehensive docstring in Google style",
-    target=function
-    context={
-        # provide additional context to the LLM
-        'usages': function.usages,
-        'dependencies': function.dependencies
-    }
-)
-function.set_docstring(new_docstring)
-```
-
 ### Documentation Coverage
 
 You can analyze and improve documentation coverage across your codebase:
@@ -200,4 +175,4 @@ for function in codebase.functions:
 
 coverage = (documented / total * 100) if total > 0 else 0
 print(f"Documentation coverage: {coverage:.1f}%")
-```
+```