"jFSSKJr SSKJrJr SSKJrJrJr SSK J r SSK J r J r JrJrJr SSKJr SSKJr SS KJr SS KJrJr SS KJr SS KJrJr SS KJ r J!r! SSK"J#r#J$r$ \ (a0SSK%J&r& SSK'J(r( SSK)J*r* SSK+J,r, SSK-J.r. SSK/J0r0 SSK1J2r2 SSK3J4r4 Sr5Sr6"SS\5r7\ "SS9"SS\\55r8"S S!\\\5r9g")#) annotations)ABCabstractmethod) AwaitableCallableSequence) dataclass) TYPE_CHECKINGAnyGenericLiteralProtocol)SchemaValidator)Self)#normalize_toolset_instruction_parts) AgentDepsT RunContext)gather)InstructionPartToolsetInstructionSource)ToolDefinitionToolsPrepareFunc)InstructionContributionmake_contributionApprovalRequiredToolsetDeferredLoadingToolsetFilteredToolsetIncludeReturnSchemasToolsetPrefixedToolsetPreparedToolsetRenamedToolsetSetMetadataToolsetzzcX\rSrSrSrSS.S SjjrSS.S SjjrSrg) SchemaValidatorProt$zsProtocol for a Pydantic Core `SchemaValidator` or `PluggableSchemaValidator` (which is private but API-compatible).F) allow_partialc gNselfinputr0kwargss Z/home/mande/repo/quber/.venv/lib/python3.13/site-packages/pydantic_ai/toolsets/abstract.py validate_json!SchemaValidatorProt.validate_json's c gr2r3r4s r8validate_python#SchemaValidatorProt.validate_python/sr;r3N)r6zstr | bytes | bytearrayr0/bool | Literal['off', 'on', 'trailing-strings']r7r returnr )r6r r0r?r7r r@r )__name__ __module__ __qualname____firstlineno____doc__r9r=__static_attributes__r3r;r8r.r.$si} JO &G    _d,[or r;r.T)kw_onlycX\rSrSr%SrS\S'S\S'S\S'S \S 'S rS \S 'Srg ) ToolsetTool4a&Definition of a tool available on a toolset. This is a wrapper around a plain tool definition that includes information about: - the toolset that provided it, for use in error messages - the maximum number of retries to attempt if the tool call fails - the validator for the tool's arguments AbstractToolset[AgentDepsT]toolsetrtool_defint max_retriesz%SchemaValidator | SchemaValidatorProtargs_validatorNzCallable[..., Any] | Noneargs_validator_funcr3)rArBrCrDrE__annotations__rQrFr3r;r8rIrI4sB)(I]J996:29r;rIc\rSrSrSr\\S!Sj55r\S"Sj5r\S"Sj5r S#Sjr S#Sjr S$Sjr S%S jr S&S jrS'S jrS'S jrS(S jrS)SjrS*SjrS+Sjr\S,Sj5r\S-Sj5rS.SjrS/SjrS0SjrS1SjrS2SjrS3SjrS4SjrS4S5SjjrS6S7SjjrS8Sjr S9Sjr!S r"g):AbstractToolsetVzA toolset is a collection of tools that can be used by an agent. It is responsible for: - Listing the tools it contains - Validating the arguments of the tools - Calling the tools See [toolset docs](../toolsets.md) for more information. c[5e)aAn ID for the toolset that is unique among all toolsets registered with the same agent. If you're implementing a concrete implementation that users can instantiate more than once, you should let them optionally pass a custom ID to the constructor and return that here. A toolset needs to have an ID in order to be used in a durable execution environment like Temporal, in which case the ID will be used to identify the toolset's activities within the workflow. IDs wrapped in angle brackets (`''` for an agent's own function toolset, `''` for its output tools) name a role the framework fills on the user's behalf rather than a registered toolset. Don't return one from your own toolset. NotImplementedErrorr5s r8idAbstractToolset.idbs "##r;czURRnUR(aUSUR<3- nU$)z2The name of the toolset for use in error messages. ) __class__rArZ)r5labels r8r_AbstractToolset.labelqs3'' 77 q _ $E r;cg)zUA hint for how to avoid name conflicts with other toolsets for use in error messages.zSRename the tool or wrap the toolset in a `PrefixedToolset` to avoid name conflicts.r3rYs r8tool_name_conflict_hint'AbstractToolset.tool_name_conflict_hintyser;c# U$7f)zReturn the toolset to use for this agent run. Called once per run, before `__aenter__`. Override this to return a fresh instance for per-run state isolation. Default: return `self` (shared across runs). r3r5ctxs r8for_runAbstractToolset.for_run~s  c# U$7f)aWReturn the toolset to use for this run step. Called at the start of each run step. Override this to return a modified instance for per-step state transitions. If returning a new instance, you are responsible for managing any lifecycle transitions (exiting old inner toolsets, entering new ones). Default: return `self` (no per-step changes). r3res r8 for_run_stepAbstractToolset.for_run_steps  ric# U$7f)zkEnter the toolset context. This is where you can set up network connections in a concrete implementation. r3rYs r8 __aenter__AbstractToolset.__aenter__s  ric # g7f)zmExit the toolset context. This is where you can tear down network connections in a concrete implementation. Nr3)r5argss r8 __aexit__AbstractToolset.__aexit__s c # g7f)apReturn instructions for how to use this toolset's tools. Override this method to provide instructions that help the agent understand how to use the tools in this toolset effectively. Simple implementations can return a plain `str`; advanced implementations can return [`InstructionPart`][pydantic_ai.messages.InstructionPart] objects to indicate whether each instruction part is static or dynamic for caching purposes. Args: ctx: The run context for this agent run. Returns: Instruction string, `InstructionPart`, list of either, or `None`. Plain `str` values are treated as dynamic instructions by default. Nr3res r8get_instructions AbstractToolset.get_instructionss &rtc# UR5(dURU5IShvN $URU5IShvN nUR5n/n[ U5HnUR bN[ UR R[5(a%URUR R5OSnUR[UbUOUU55 M U$NN7f)aCollect contributions once, preserving the toolset that authored every relayed part. A toolset that only passes its children along is walked; anything that speaks for itself is asked once. That covers a leaf and a container whose subclass took `get_instructions` over with the same path, because the difference between them is only what they own: a returned key owned below is relayed unchanged and stays attributed to its owner, and everything else is the caller's own text, resolved against the caller's own key. N) _authors_own_instructions(_collect_child_instruction_contributionsrv_instruction_sources_by_keyrrZ isinstancesourcergetappendr)r5rfresultsources_by_key contributionspartowners r8"_collect_instruction_contributions2AbstractToolset._collect_instruction_contributionss--//FFsKK K,,S1199;CE 7?D 77&:dggnnF^+_+_""477>>2   !2E# UHoRT5v M g7fr2)r).0childrfs r8 KAbstractToolset._collect_child_instruction_contributions..s fIe66s;;Ies N)r_instruction_children)r5rfchild_contributionsr contributions ` r8rz8AbstractToolset._collect_child_instruction_contributionssX%+ fIcIcIe f%  3Fh2EZg, Zg 2Ehh is $AA  AA A Acg)zCThe toolsets whose instruction contributions this one passes along.r3r3rYs r8r%AbstractToolset._instruction_childrensr;cg)aaWhether `get_instructions` speaks for this toolset rather than aggregating its children. True here because a toolset with nothing below it can only be speaking for itself. A container overrides this to answer for the case that actually varies: whether a subclass has taken the method over, or it is still the inherited implementation that just relays. Tr3rYs r8ry)AbstractToolset._authors_own_instructionssr;chURbSUR;ag[UR5$)zURead this toolset's source without validating an id that contributes no instructions.N:)rZrrYs r8_instruction_source#AbstractToolset._instruction_sources' 77?cTWWn'00r;c0nUR5H;nUR5R5Hup4URX45 M M= UR 5=n(aURX@5 U$)a]Map every source key at or below this toolset to the toolset that owns it. Children are inserted first and the container is inserted last without overwriting them, so a child remains the owner when a malformed tree repeats its key at a container boundary. The duplicate contribution check reports the ambiguity if both sources actually contribute. )rr{items setdefaultr)r5sourcesr source_idr}s r8r{+AbstractToolset._instruction_sources_by_keysuPR//1E%*%F%F%H%N%N%P! ""95&Q2--/ /6 /   v ,r;c# [5e7f)z-The tools that are available in this toolset.rWres r8 get_toolsAbstractToolset.get_toolss"## c# [5e7f)aCall a tool with the given arguments. Args: name: The name of the tool to call. tool_args: The arguments to pass to the tool. ctx: The run context. tool: The tool definition returned by [`get_tools`][pydantic_ai.toolsets.AbstractToolset.get_tools] that was called. rW)r5name tool_argsrftools r8 call_toolAbstractToolset.call_tools"##rcZ# URU5IShvN UR$N7f)uReturn the tool to call for a tool definition this toolset already produced. Used by [durable execution](../durable_execution/overview.md) to rebuild the tool inside a durable unit from the definition a discovery unit already recorded, instead of listing the toolset's tools a second time. The default lists them, which is always correct; a toolset that can build the tool from the definition alone — like [`MCPToolset`][pydantic_ai.mcp.MCPToolset], whose listing is a network round trip — should override this to skip the listing. Args: tool_def: The tool definition to build the tool from. ctx: The run context. Raises: KeyError: If this toolset holds no tool under that name. N)rr)r5rMrfs r8get_tool_for_tool_def%AbstractToolset.get_tool_for_tool_defs%&nnS))8==99)s +)+cU"U5 g)zmRun a visitor function on all "leaf" toolsets (i.e. those that implement their own tool listing and calling).Nr3r5visitors r8applyAbstractToolset.apply#s  r;cU"U5$)zRun a visitor function on all "leaf" toolsets (i.e. those that implement their own tool listing and calling) and replace them in the hierarchy with the result of the function.r3rs r8visit_and_replace!AbstractToolset.visit_and_replace'st}r;cSSKJn U"X5$)zReturns a new toolset that filters this toolset's tools using a filter function that takes the agent context and the tool definition. See [toolset docs](../toolsets.md#filtering-tools) for more information. rr!)filteredr")r5 filter_funcr"s r8rAbstractToolset.filtered-s .t11r;cSSKJn U"X5$)zReturns a new toolset that prefixes the names of this toolset's tools. See [toolset docs](../toolsets.md#prefixing-tool-names) for more information. rr%)prefixedr&)r5prefixr&s r8rAbstractToolset.prefixed8s .t,,r;cSSKJn U"X5$)zReturns a new toolset that prepares this toolset's tools using a prepare function that takes the agent context and the original tool definitions. See [toolset docs](../toolsets.md#preparing-tool-definitions) for more information. rr')preparedr()r5 prepare_funcr(s r8rAbstractToolset.preparedAs .t22r;cSSKJn U"X5$)zReturns a new toolset that renames this toolset's tools using a dictionary mapping new names to original names. See [toolset docs](../toolsets.md#renaming-tools) for more information. rr))renamedr*)r5name_mapr*s r8rAbstractToolset.renamedJs ,d--r;cg)NTr3)rfrMrs r8AbstractToolset.VsTr;cSSKJn U"X5$)zReturns a new toolset that requires (some) calls to tools it contains to be approved. See [toolset docs](../toolsets.md#requiring-tool-approval) for more information. rr)approval_requiredr)r5approval_required_funcrs r8r!AbstractToolset.approval_requiredSs ?&tDDr;Nc:SSKJn U"Xb [U5S9$SS9$)aReturns a new toolset that marks tools for deferred loading, hiding them until revealed. Tool search, `load_capability` and another tool's `ToolReturn.tools` all reveal. See [toolset docs](../toolsets.md#deferred-loading) for more information. Args: tool_names: Optional sequence of tool names to mark for deferred loading. If `None`, all tools are marked for deferred loading. rrN) tool_names)deferred_loadingr frozenset)r5rr s r8 defer_loadingAbstractToolset.defer_loadingas% =%dPfy7Lqqlpqqr;cSSKJn U"U5$)arReturns a new toolset that sets `include_return_schema=True` on all tools. This causes the model to receive return type information for the tools in this toolset. For models that natively support return schemas (e.g. Google Gemini), the schema is passed as a structured field. For other models, it is injected into the tool description as JSON text. This is the toolset-level equivalent of the [`IncludeToolReturnSchemas`][pydantic_ai.capabilities.IncludeToolReturnSchemas] capability, which can be used to enable return schemas across all toolsets or a subset matched by a [`ToolSelector`][pydantic_ai.tools.ToolSelector]. rr#)include_return_schemasr$)r5r$s r8r&AbstractToolset.include_return_schemasps H*400r;c SSKJn U"X5$)zDReturns a new toolset that merges the given metadata onto all tools.rr+) set_metadatar,)r5metadatar,s r8 with_metadataAbstractToolset.with_metadatas4!$11r;r3)r@z str | None)r@str)rfRunContext[AgentDepsT]r@rK)r@r)rqr r@z bool | None)rfrr@z>str | InstructionPart | Sequence[str | InstructionPart] | None)rfrr@z)list[InstructionContribution[AgentDepsT]])r@z%Sequence[AbstractToolset[AgentDepsT]])r@bool)r@zToolsetInstructionSource | None)r@z;dict[ToolsetInstructionSource, AbstractToolset[AgentDepsT]])rfrr@z"dict[str, ToolsetTool[AgentDepsT]]) rrrzdict[str, Any]rfrrToolsetTool[AgentDepsT]r@r )rMrrfrr@r)rz-Callable[[AbstractToolset[AgentDepsT]], None]r@None)rzDCallable[[AbstractToolset[AgentDepsT]], AbstractToolset[AgentDepsT]]r@rK)rzJCallable[[RunContext[AgentDepsT], ToolDefinition], bool | Awaitable[bool]]r@zFilteredToolset[AgentDepsT])rrr@zPrefixedToolset[AgentDepsT])rzToolsPrepareFunc[AgentDepsT]r@zPreparedToolset[AgentDepsT])rzdict[str, str]r@zRenamedToolset[AgentDepsT])rzHCallable[[RunContext[AgentDepsT], ToolDefinition, dict[str, Any]], bool]r@z#ApprovalRequiredToolset[AgentDepsT]r2)rzSequence[str] | Noner@z"DeferredLoadingToolset[AgentDepsT])r@z'IncludeReturnSchemasToolset[AgentDepsT])rr r@zSetMetadataToolset[AgentDepsT])#rArBrCrDrEpropertyrrZr_rbrgrkrnrrrvrrzrryrr{rrrrrrrrrrrrrrFr3r;r8rTrTVs  $ $ee) G*) 2:i)i 2i1 $$ $ $$2 $9O $Wn $  $ $:&:-C: :*[ $ 2e 2 $ 2-3. 2 E h E - E r1$2r;rTN): __future__rabcrrcollections.abcrrr dataclassesr typingr r r r r pydantic_corertyping_extensionsr _instructionsr _run_contextrr_utilsrmessagesrrtoolsrr_instruction_collectionrrrrrr rr"rr$rr&rr(rr*rr,AGENT_TOOLSET_IDOUTPUT_TOOLSET_IDr.rIrTr3r;r8rs"#99!AA)"?1@4O:8)C))'0tn (   4'*%Bp2c7:.p2r;