tools
A Claude Code settings key, read out of the shipped bundle. It has been in the build since v2.1.284, including the newest one read.
Tool discovery uses a newer path with retries on servers that negotiate the modern protocol; older servers are unchanged.
Not Anthropic's. This is the line from the earliest changelog entry here that named it, v2.1.221.
In the changelogs
26Releases whose published page names tools.
-
v2.1.285
Tool restrictions and turn limits now carry over to spawned sessions
Spawned sessions now keep your
disallowedToolsandtoolsrestrictions and yourmaxTurnslimitfound in this entry's text
-
v2.1.282
Post-turn memory config accepts an extract_transport choice
The post-turn memory config now takes an extract_transport option, "tools" or "text", defaulting to "tools"
found in this entry's text
-
v2.1.281
SDK can hand in-process MCP server manifests to the CLI at initialize
The Agent SDK can now read tool lists from in-process MCP servers itself and pass them to the CLI when it starts
found in this entry's text
-
v2.1.274
Internal: session options gain a shared
toolCatalogobjectInternal: sessions now pass around a shared toolCatalog object instead of calling tool-lookup functions directly
found in this entry's text
-
v2.1.271
mcp serve now advertises and sends tools/list_changed notifications
claude mcp serve now tells connected clients when its tool list changes
found in this entry's text
-
v2.1.269
'Tether' thread-continuation requests can omit system prompt/tools when continuing, behind two off-by-default flags, with a fingerprint-mismatch circuit breaker
Continuing a server-side conversation thread can now skip resending the system prompt and tools list, with a fallback if the server disagrees
found in this entry's text
-
v2.1.269
Model-parameter allowlist changed from array to Set
Internal list of allowed API request fields switched from an array to a Set
found in this entry's text
-
v2.1.267
New tool_schema_invalid error classification for 400s referencing tools.N.input_schema
API 400 errors about an invalid tool input schema now get their own error type instead of being generic
found in this entry's text
-
v2.1.265
New "unknown_tools" error case for artifact MCP manifests
Publishing an artifact now fails clearly if its MCP manifest lists tool names the connector doesn't actually have
found in this entry's text
-
v2.1.265
SDK MCP servers can now skip the control-channel handshake via manifests
SDK-registered in-process MCP servers can now skip a round-trip handshake at session start
found in this entry's text
-
v2.1.260
SDK init frame now reports skills and plugins in use
SDK init handshake now reports which skills and plugins are in use.
found in this entry's text
-
v2.1.260
hostPrompt field threaded through plugin/skill tool schema
A new hostPrompt field is now threaded through plugin/skill tool schema building.
found in this entry's text
-
v2.1.259
MCP retry loops for tools/list and generic retry helper stop early on abort
Cancelled MCP retries now stop during the backoff window instead of continuing through it.
found in this entry's text
-
v2.1.257
MCP client reconnect state merge moved into a shared helper
Merging a reconnected MCP server's tools, commands and resources moved into one shared helper.
found in this entry's text
-
v2.1.257
Artifact tool split into Comments/Data/Check sub-tools
Artifact actions are now attributed to three separate tools, ArtifactComments, ArtifactData and ArtifactCheck, in permission rules.
found in this entry's text
-
v2.1.248
Missing MCP capabilities return empty instead of erroring
MCP servers that advertise no tools or prompts return empty instead of erroring.
found in this entry's text
-
v2.1.242
Agent-backed mock responders are present but switched off
Mocks could be answered by a prompt instead of a fixed body, but that type is rejected.
found in this entry's text
-
v2.1.239
Remote tool execution is complete but hard-disabled in this build
Remote device execution is fully built but hard-disabled; no tool can be sent to a device.
found in this entry's text
-
v2.1.235
Cloud/frame SDK messages are validated instead of trusted
Cloud/frame SDK messages are now validated field-by-field instead of trusted
found in this entry's text
-
v2.1.224
vault_idsdocumented as create-only, and deployments can be updatedDocs say vault_ids cannot change mid-session, and deployments now have an update endpoint.
found in this entry's text
-
v2.1.221
tools/list uses the new client listTools path with retry backoff on modern servers
Tool discovery uses a newer path with retries on servers that negotiate the modern protocol; older servers are unchanged.
found in this entry's text
-
v2.1.221
Managed MCP servers carry discover support and a cached discover response
Managed MCP server configs now carry discovery data and strip more server-supplied junk before writing to your config files.
found in this entry's text
-
v2.1.208
Official plugin prompt overrides via server-side feature flag
found in this entry's text
-
v2.1.133
Skill Tool Guidance Updated
found in this entry's text
-
v2.1.30
Enhanced Custom Subagent Schema
found in this entry's text
-
v2.0.50
MCP CLI Endpoint
found in this entry's text
First cited
0No release's published evidence quotes this name. The ledger indexes the Evidence lines of every changelog on this site, so this says nobody here has ever quoted it; it says nothing about how old the name is.
Presence across releases
1Build by build
2One row per release since the first build this name was read out of. Absent means the build was read and the name was not in it, never mined means no bundle for that release was ever archived, and a declared type or a default is only ever what that release's own bundle stated.
Read out of the published npm bundle release by release, and out of Anthropic's own documentation as this site captured it. Nothing on this page is a description anybody here wrote about what the settings key does. All settings keys.