Repository navigation
feat: Keep the Tool Set Fixed per Segment with Tool Groups and the SDK Dispatcher - #17147
Open
Shabonasar wants to merge 91 commits into
Open
Shabonasar wants to merge 91 commits into
Shabonasar wants to merge 91 commits into
Conversation
The per-request web search context (current date/time with milliseconds) was joined before additional_instructions, which carry mostly static text such as the artifacts prompt. Provider-side prefix caches stopped at the timestamp, so the static text after it was recomputed on every request. Move the dynamic context to the end via buildAdditionalInstructions and cover the ordering with unit tests.
…che-order # Conflicts: # packages/api/src/agents/initialize.ts
…e date changes in history The web search runtime context and temporal special variables were anchored to the turn start with millisecond precision. The system message precedes the whole history, so every turn invalidated the cached history and new conversations could not reuse the system prompt (LibreChat-AI#16877). - Web search runtime context states the conversation start in the user's timezone with minute precision; temporal variables resolve against the same persisted anchor. - A user message sent on a later local date than the model last saw gets a <context_update> block in its prompt copy only, derived from stored createdAt values so it renders identically on later turns; compaction boundaries follow the agents SDK formatter. - Optional endpoints.agents.promptMessageTimestamps adds the send time to every user message; the turn's user message is stored with createdAt = turnStartedAt on every save path. - Provider payload stability tests through createRun for OpenAI, an OpenAI-compatible endpoint, Anthropic, OpenRouter, Bedrock and Google.
… routes The browser resolved promptPrefix with its own clock on every send, so turn 1 of a spec-backed conversation carried the raw template while turn 2 carried resolved text in a different system-prompt position. Only Assistants routes resolve on the client now; the server resolves against the conversation start.
… of a conversation
Adds `endpoints.agents.promptCache.stableContext` (default off). Off, the prompt is
built as before. On:
- prompt time (`{{current_datetime}}` and friends, the web search runtime context) is
anchored to the conversation start, and a date change is announced by a
`<context_update>` block at the start of the first user message sent after it;
- the conversation-stable part of `additional_instructions` is merged into the system
instructions for Anthropic and OpenRouter with prompt caching, while memory, RAG and
one-turn queued file context stay in `additional_instructions`, where the SDK moves
them next to the newest user message;
- dynamic tool context goes after the additional instructions;
- the user message is inserted with the turn start as `createdAt`, so later turns render
the same stamp.
`promptMessageTimestamps` (default off) additionally prefixes each user message with the
time it was sent.
The history notes, the instruction split and the lever live in `packages/api`; the
controller only calls them. `applyContextToAgent` accepts the run context as parts and
records which are volatile, so no text is split by matching.
…assified parts Producers (project, memory guard, artifacts, skills catalog, RAG, memory, agent context files, always-apply skills of isolated subagents) contribute keyed parts through contributeContext instead of concatenating strings into additional_instructions. One registry fixes the canonical order and classifies every part, tool context map entries included, as static, conversation or turn. A runtime key the registry does not know counts as turn, so a new producer cannot break the stable prefix by default. assembleSystemPrompt builds the strings handed to the SDK. Once parts exist they are the only source of the tail; a direct write to additional_instructions is ignored with a warning. Without promptCache.stableContext it reproduces the previous request; with it, createRun moves only the turn parts with the prompt-cache tail and merges the rest into the instructions. The assembled text is byte-identical to string concatenation for both values of the lever, guarded by a snapshot recorded on the previous implementation. Isolated subagents with an always-apply skill place the skill ahead of the tool runtime context when stableContext is on.
…m prompt per segment Behind `endpoints.agents.promptCache.stableContext`; off, nothing here runs and the prompt is built as before. A single send-time step (turnContext) decides what the primary agent's model sees. Context sources (project, artifacts mode, memory tool guard, skills catalog, tool runtime context, memory, date, send time, retrieved context, queued upload files) are sampled once when the user sends a message; only the net difference from what the model already knows along the parent chain is stored on that user message (message.contextUpdates) and rendered as one <context_update> block at the start of its prompt copy. Toggles flipped back and forth before sending leave no trace, and stored values render the same bytes on every later turn, whatever the timezone. A regeneration sends no message: it renders what the message stored and writes nothing. The system prompt is rendered from the segment's anchor instead of the live parts, so it stays byte-identical until the next compaction. The anchor of the first segment is stored on the conversation, the anchor of a segment opened by a compaction on the message carrying the summary (contextAnchor); a conversation without one is anchored to its first turn after the upgrade. A re-anchor keeps the previous order, drops what is off and appends what is new, and dates itself, so the first message after the boundary announces nothing. Memory is a state source: an update lists the entries added, changed and removed (by key where the memory has keys) instead of the whole memory, and no longer rewrites the system prompt; memory switched off is withheld from the whole request. Retrieved context is stored with the user message it was retrieved for, like the file context bound to the message that carried the files; isolated subagents still take it in their system prompt. Queued upload files describe their own message only. A resumed turn (tool approval or an answered question) renders its system prompt from the paused turn's anchor, not from the live state. The planning, the anchor storage and the token accounting live in packages/api; the controller calls them. The conversation anchor is read with the conversation the ingress already loads, and a resume reads the messages once with a projection. contextUpdates and contextAnchor are server-written only, select: false, and stripped from client input, so they stay out of reads, shares and forks.
…d the SDK dispatcher Behind `endpoints.agents.promptCache.stableContext`, which also stays off, with one warning, while the installed @librechat/agents has no tool dispatcher. Providers serialize `tools` ahead of the system prompt, so switching an MCP server, an action or an agent tool mid-conversation reset the prompt cache from position 0. - Tool groups (`tool:mcp:<server>`, `tool:action:<domain>`, `tool:agent:<toolkit>`) are a turn-context source. The segment anchor stores each group's schemas and MCP instructions; `tools` and the MCP instructions are rendered from it until the next compaction. - The initial set orders MCP servers and their tools by name, other groups as the agent lists them, after the ungrouped tools. A compaction keeps the anchored order, drops groups that are off and appends the new ones. - A group enabled mid-segment is deferred and announced once in the user message with its schemas; the model calls it through the SDK dispatcher (`toolDispatch`), which resolves the call before hooks, approval and remembered approvals, so they see the real tool. An anchored group that is off keeps its schemas with `disabled`; an unreachable one is `unavailable` without an update. - A schema a server changes mid-segment is announced once, in the message sent with the change: the changed parameters, tools that are new (reached through the dispatcher) and tools that are gone. `tools` itself changes at the next anchor. - Tools found by `tool_search` stay deferred and are called through the dispatcher; the host search returns their schemas. - The formatter tool set keeps past dispatcher calls; isolated subagent copies bind the tools as loaded. - MCP servers and their tools load in name order, and MCP instructions are listed by server name; both only with the lever. The instructions of the named servers are read in packages/api, so MCPManager keeps its API. - UI: `resolveDisplayToolCall` routes a dispatcher call to the card and the group header of the tool it ran; the stream reducer keeps `dispatched`. The package depends on the tool dispatcher of @librechat/agents, which is not in a release yet; `package.json` moves to the release that has it.
This was referenced Oct 15, 2026
codegraph-librechat
Bot
added
🗺️ Backend Infra
Shabonasar
marked this pull request as draft
October 16, 2026 06:21
…assified parts Producers (project, memory guard, artifacts, skills catalog, RAG, memory, agent context files, always-apply skills of isolated subagents) contribute keyed parts through contributeContext instead of concatenating strings into additional_instructions. One registry fixes the canonical order and classifies every part, tool context map entries included, as static, conversation or turn. A runtime key the registry does not know counts as turn, so a new producer cannot break the stable prefix by default. assembleSystemPrompt builds the strings handed to the SDK. Once parts exist they are the only source of the tail; a direct write to additional_instructions is ignored with a warning. Without promptCache.stableContext it reproduces the previous request; with it, createRun moves only the turn parts with the prompt-cache tail and merges the rest into the instructions. The assembled text is byte-identical to string concatenation for both values of the lever, guarded by a snapshot recorded on the previous implementation. Isolated subagents with an always-apply skill place the skill ahead of the tool runtime context when stableContext is on.
…m prompt per segment Behind `endpoints.agents.promptCache.stableContext`; off, nothing here runs and the prompt is built as before. A single send-time step (turnContext) decides what the primary agent's model sees. Context sources (project, artifacts mode, memory tool guard, skills catalog, tool runtime context, memory, date, send time, retrieved context, queued upload files) are sampled once when the user sends a message; only the net difference from what the model already knows along the parent chain is stored on that user message (message.contextUpdates) and rendered as one <context_update> block at the start of its prompt copy. Toggles flipped back and forth before sending leave no trace, and stored values render the same bytes on every later turn, whatever the timezone. A regeneration sends no message: it renders what the message stored and writes nothing. The system prompt is rendered from the segment's anchor instead of the live parts, so it stays byte-identical until the next compaction. The anchor of the first segment is stored on the conversation, the anchor of a segment opened by a compaction on the message carrying the summary (contextAnchor); a conversation without one is anchored to its first turn after the upgrade. A re-anchor keeps the previous order, drops what is off and appends what is new, and dates itself, so the first message after the boundary announces nothing. Memory is a state source: an update lists the entries added, changed and removed (by key where the memory has keys) instead of the whole memory, and no longer rewrites the system prompt; memory switched off is withheld from the whole request. Retrieved context is stored with the user message it was retrieved for, like the file context bound to the message that carried the files; isolated subagents still take it in their system prompt. Queued upload files describe their own message only. A resumed turn (tool approval or an answered question) renders its system prompt from the paused turn's anchor, not from the live state. The planning, the anchor storage and the token accounting live in packages/api; the controller calls them. The conversation anchor is read with the conversation the ingress already loads, and a resume reads the messages once with a projection. contextUpdates and contextAnchor are server-written only, select: false, and stripped from client input, so they stay out of reads, shares and forks.
…d the SDK dispatcher Behind `endpoints.agents.promptCache.stableContext`, which also stays off, with one warning, while the installed @librechat/agents has no tool dispatcher. Providers serialize `tools` ahead of the system prompt, so switching an MCP server, an action or an agent tool mid-conversation reset the prompt cache from position 0. - Tool groups (`tool:mcp:<server>`, `tool:action:<domain>`, `tool:agent:<toolkit>`) are a turn-context source. The segment anchor stores each group's schemas and MCP instructions; `tools` and the MCP instructions are rendered from it until the next compaction. - The initial set orders MCP servers and their tools by name, other groups as the agent lists them, after the ungrouped tools. A compaction keeps the anchored order, drops groups that are off and appends the new ones. - A group enabled mid-segment is deferred and announced once in the user message with its schemas; the model calls it through the SDK dispatcher (`toolDispatch`), which resolves the call before hooks, approval and remembered approvals, so they see the real tool. An anchored group that is off keeps its schemas with `disabled`; an unreachable one is `unavailable` without an update. - A schema a server changes mid-segment is announced once, in the message sent with the change: the changed parameters, tools that are new (reached through the dispatcher) and tools that are gone. `tools` itself changes at the next anchor. - Tools found by `tool_search` stay deferred and are called through the dispatcher; the host search returns their schemas. - The formatter tool set keeps past dispatcher calls; isolated subagent copies bind the tools as loaded. - MCP servers and their tools load in name order, and MCP instructions are listed by server name; both only with the lever. The instructions of the named servers are read in packages/api, so MCPManager keeps its API. - UI: `resolveDisplayToolCall` routes a dispatcher call to the card and the group header of the tool it ran; the stream reducer keeps `dispatched`. The package depends on the tool dispatcher of @librechat/agents, which is not in a release yet; `package.json` moves to the release that has it.
Shabonasar
marked this pull request as ready for review
October 16, 2026 17:20
The per-request web search context (current date/time with milliseconds) was joined before additional_instructions, which carry mostly static text such as the artifacts prompt. Provider-side prefix caches stopped at the timestamp, so the static text after it was recomputed on every request. Move the dynamic context to the end via buildAdditionalInstructions and cover the ordering with unit tests.
… of a conversation
Adds `endpoints.agents.promptCache.stableContext` (default off). Off, the prompt is
built as before. On:
- prompt time (`{{current_datetime}}` and friends, the web search runtime context) is
anchored to the conversation start, and a date change is announced by a
`<context_update>` block at the start of the first user message sent after it;
- the conversation-stable part of `additional_instructions` is merged into the system
instructions for Anthropic and OpenRouter with prompt caching, while memory, RAG and
one-turn queued file context stay in `additional_instructions`, where the SDK moves
them next to the newest user message;
- dynamic tool context goes after the additional instructions;
- the user message is inserted with the turn start as `createdAt`, so later turns render
the same stamp.
`promptMessageTimestamps` (default off) additionally prefixes each user message with the
time it was sent.
The history notes, the instruction split and the lever live in `packages/api`; the
controller only calls them. `applyContextToAgent` accepts the run context as parts and
records which are volatile, so no text is split by matching.
…m prompt per segment Behind `endpoints.agents.promptCache.stableContext`; off, nothing here runs and the prompt is built as before. A single send-time step (turnContext) decides what the primary agent's model sees. Context sources (project, artifacts mode, memory tool guard, skills catalog, tool runtime context, memory, date, send time, retrieved context, queued upload files) are sampled once when the user sends a message; only the net difference from what the model already knows along the parent chain is stored on that user message (message.contextUpdates) and rendered as one <context_update> block at the start of its prompt copy. Toggles flipped back and forth before sending leave no trace, and stored values render the same bytes on every later turn, whatever the timezone. A regeneration sends no message: it renders what the message stored and writes nothing. The system prompt is rendered from the segment's anchor instead of the live parts, so it stays byte-identical until the next compaction. The anchor of the first segment is stored on the conversation, the anchor of a segment opened by a compaction on the message carrying the summary (contextAnchor); a conversation without one is anchored to its first turn after the upgrade. A re-anchor keeps the previous order, drops what is off and appends what is new, and dates itself, so the first message after the boundary announces nothing. Memory is a state source: an update lists the entries added, changed and removed (by key where the memory has keys) instead of the whole memory, and no longer rewrites the system prompt; memory switched off is withheld from the whole request. Retrieved context is stored with the user message it was retrieved for, like the file context bound to the message that carried the files; isolated subagents still take it in their system prompt. Queued upload files describe their own message only. A resumed turn (tool approval or an answered question) renders its system prompt from the paused turn's anchor, not from the live state. The planning, the anchor storage and the token accounting live in packages/api; the controller calls them. The conversation anchor is read with the conversation the ingress already loads, and a resume reads the messages once with a projection. contextUpdates and contextAnchor are server-written only, select: false, and stripped from client input, so they stay out of reads, shares and forks.
…assified parts Producers (project, memory guard, artifacts, skills catalog, RAG, memory, agent context files, always-apply skills of isolated subagents) contribute keyed parts through contributeContext instead of concatenating strings into additional_instructions. One registry fixes the canonical order and classifies every part, tool context map entries included, as static, conversation or turn. A runtime key the registry does not know counts as turn, so a new producer cannot break the stable prefix by default. assembleSystemPrompt builds the strings handed to the SDK. Once parts exist they are the only source of the tail; a direct write to additional_instructions is ignored with a warning. Without promptCache.stableContext it reproduces the previous request; with it, createRun moves only the turn parts with the prompt-cache tail and merges the rest into the instructions. The assembled text is byte-identical to string concatenation for both values of the lever, guarded by a snapshot recorded on the previous implementation. Isolated subagents with an always-apply skill place the skill ahead of the tool runtime context when stableContext is on.
A handoff target that shares the primary agent's memory partition kept a live copy of the memory in its per-turn tail. With prompt caching the SDK relocates that tail next to the newest user message, so the memory sat at a different place on every turn: the target's earlier requests no longer matched its next one, which missed the cache and, for models that bind thinking blocks to the prefix, made the API drop the target's own thinking blocks. The memory the turn context anchors for the segment now goes into the target's conversation block, and its changes reach it as the <context_update> blocks already stored on the history messages, as they reach the primary agent. Targets on another memory partition keep their live copy.
…he runtime note stays byte-stable The code-file note of an attached workspace is runtime context that the turn context anchors and updates. Priming returns files by last update, which moves with usage, so the same files could be listed in another order on a later turn and the note was announced again without having changed.
…d the SDK dispatcher Behind `endpoints.agents.promptCache.stableContext`, which also stays off, with one warning, while the installed @librechat/agents has no tool dispatcher. Providers serialize `tools` ahead of the system prompt, so switching an MCP server, an action or an agent tool mid-conversation reset the prompt cache from position 0. - Tool groups (`tool:mcp:<server>`, `tool:action:<domain>`, `tool:agent:<toolkit>`) are a turn-context source. The segment anchor stores each group's schemas and MCP instructions; `tools` and the MCP instructions are rendered from it until the next compaction. - The initial set orders MCP servers and their tools by name, other groups as the agent lists them, after the ungrouped tools. A compaction keeps the anchored order, drops groups that are off and appends the new ones. - A group enabled mid-segment is deferred and announced once in the user message with its schemas; the model calls it through the SDK dispatcher (`toolDispatch`), which resolves the call before hooks, approval and remembered approvals, so they see the real tool. An anchored group that is off keeps its schemas with `disabled`; an unreachable one is `unavailable` without an update. - A schema a server changes mid-segment is announced once, in the message sent with the change: the changed parameters, tools that are new (reached through the dispatcher) and tools that are gone. `tools` itself changes at the next anchor. - Tools found by `tool_search` stay deferred and are called through the dispatcher; the host search returns their schemas. - The formatter tool set keeps past dispatcher calls; isolated subagent copies bind the tools as loaded. - MCP servers and their tools load in name order, and MCP instructions are listed by server name; both only with the lever. The instructions of the named servers are read in packages/api, so MCPManager keeps its API. - UI: `resolveDisplayToolCall` routes a dispatcher call to the card and the group header of the tool it ran; the stream reducer keeps `dispatched`. The package depends on the tool dispatcher of @librechat/agents, which is not in a release yet; `package.json` moves to the release that has it.
…ed tools and instructions A librechat.yaml MCP server hidden from the acting user by a config profile is an access boundary: its instructions must not reach the model and its tools must not load. An anchored group of such a server stayed bound from its stored schemas, with its instructions in the stable system prompt, until the next anchor. - `LiveToolGroups.hidden` holds the group keys of the servers hidden from the user, read where the groups are sampled through an injectable reader (`getHiddenMCPServerNames`). The raw names are normalized like the group keys. A failed read is not swallowed: hiding fails closed upstream, and a turn that cannot tell what is hidden must not bind its schemas. - The tool plan leaves an anchored MCP group out, schemas and instructions, when its server is hidden and not loaded. A loaded group with a hidden name is a user-sourced server and stays.
…memory Since LibreChat-AI#16301 formatMemoryContext('') returns the bare memory instructions sentence, and prepareRuntimeAgent applied it to the primary agent whenever the user had no saved memories. An agent with no memory tools and no automatic memory extraction can neither read nor write memory, so the sentence told the model nothing it could act on, yet it reached the model as per-turn context. The empty-store notice is now added only when the agent has memory tools or automatic extraction is on. Saved memories are still injected as before.
…m prompt per segment Behind `endpoints.agents.promptCache.stableContext`; off, nothing here runs and the prompt is built as before. A single send-time step (turnContext) decides what the primary agent's model sees. Context sources (project, artifacts mode, memory tool guard, skills catalog, tool runtime context, memory, date, send time, retrieved context, queued upload files) are sampled once when the user sends a message; only the net difference from what the model already knows along the parent chain is stored on that user message (message.contextUpdates) and rendered as one <context_update> block at the start of its prompt copy. Toggles flipped back and forth before sending leave no trace, and stored values render the same bytes on every later turn, whatever the timezone. A regeneration sends no message: it renders what the message stored and writes nothing. The system prompt is rendered from the segment's anchor instead of the live parts, so it stays byte-identical until the next compaction. The anchor of the first segment is stored on the conversation, the anchor of a segment opened by a compaction on the message carrying the summary (contextAnchor); a conversation without one is anchored to its first turn after the upgrade. A re-anchor keeps the previous order, drops what is off and appends what is new, and dates itself, so the first message after the boundary announces nothing. Memory is a state source: an update lists the entries added, changed and removed (by key where the memory has keys) instead of the whole memory, and no longer rewrites the system prompt; memory switched off is withheld from the whole request. Retrieved context is stored with the user message it was retrieved for, like the file context bound to the message that carried the files; isolated subagents still take it in their system prompt. Queued upload files describe their own message only. A resumed turn (tool approval or an answered question) renders its system prompt from the paused turn's anchor, not from the live state. The planning, the anchor storage and the token accounting live in packages/api; the controller calls them. The conversation anchor is read with the conversation the ingress already loads, and a resume reads the messages once with a projection. contextUpdates and contextAnchor are server-written only, select: false, and stripped from client input, so they stay out of reads, shares and forks.
A handoff target that shares the primary agent's memory partition kept a live copy of the memory in its per-turn tail. With prompt caching the SDK relocates that tail next to the newest user message, so the memory sat at a different place on every turn: the target's earlier requests no longer matched its next one, which missed the cache and, for models that bind thinking blocks to the prefix, made the API drop the target's own thinking blocks. The memory the turn context anchors for the segment now goes into the target's conversation block, and its changes reach it as the <context_update> blocks already stored on the history messages, as they reach the primary agent. Targets on another memory partition keep their live copy.
…he runtime note stays byte-stable The code-file note of an attached workspace is runtime context that the turn context anchors and updates. Priming returns files by last update, which moves with usage, so the same files could be listed in another order on a later turn and the note was announced again without having changed.
…se memory The send snapshot formatted an empty memory store as the bare memory instructions sentence whatever the agent could do with it, so the turn context anchored a sentence that tells the model nothing it can act on. The snapshot now applies the rule of prepareRuntimeAgent: an empty store is announced only to an agent with memory tools or when automatic extraction is on. Saved memories are anchored as before.
…d the SDK dispatcher Behind `endpoints.agents.promptCache.stableContext`, which also stays off, with one warning, while the installed @librechat/agents has no tool dispatcher. Providers serialize `tools` ahead of the system prompt, so switching an MCP server, an action or an agent tool mid-conversation reset the prompt cache from position 0. - Tool groups (`tool:mcp:<server>`, `tool:action:<domain>`, `tool:agent:<toolkit>`) are a turn-context source. The segment anchor stores each group's schemas and MCP instructions; `tools` and the MCP instructions are rendered from it until the next compaction. - The initial set orders MCP servers and their tools by name, other groups as the agent lists them, after the ungrouped tools. A compaction keeps the anchored order, drops groups that are off and appends the new ones. - A group enabled mid-segment is deferred and announced once in the user message with its schemas; the model calls it through the SDK dispatcher (`toolDispatch`), which resolves the call before hooks, approval and remembered approvals, so they see the real tool. An anchored group that is off keeps its schemas with `disabled`; an unreachable one is `unavailable` without an update. - A schema a server changes mid-segment is announced once, in the message sent with the change: the changed parameters, tools that are new (reached through the dispatcher) and tools that are gone. `tools` itself changes at the next anchor. - Tools found by `tool_search` stay deferred and are called through the dispatcher; the host search returns their schemas. - The formatter tool set keeps past dispatcher calls; isolated subagent copies bind the tools as loaded. - MCP servers and their tools load in name order, and MCP instructions are listed by server name; both only with the lever. The instructions of the named servers are read in packages/api, so MCPManager keeps its API. - UI: `resolveDisplayToolCall` routes a dispatcher call to the card and the group header of the tool it ran; the stream reducer keeps `dispatched`. The package depends on the tool dispatcher of @librechat/agents, which is not in a release yet; `package.json` moves to the release that has it.
…ed tools and instructions A librechat.yaml MCP server hidden from the acting user by a config profile is an access boundary: its instructions must not reach the model and its tools must not load. An anchored group of such a server stayed bound from its stored schemas, with its instructions in the stable system prompt, until the next anchor. - `LiveToolGroups.hidden` holds the group keys of the servers hidden from the user, read where the groups are sampled through an injectable reader (`getHiddenMCPServerNames`). The raw names are normalized like the group keys. A failed read is not swallowed: hiding fails closed upstream, and a turn that cannot tell what is hidden must not bind its schemas. - The tool plan leaves an anchored MCP group out, schemas and instructions, when its server is hidden and not loaded. A loaded group with a hidden name is a user-sourced server and stays.
No file changes (tree of dev c81e244). Joins earlier snapshots of the series that the branches share, so any two of them have a single merge base and merge cleanly.
With endpoints.agents.promptCache.stableContext, prompt time is anchored to the conversation start, a later date is announced in the history, the attached workspace's attachment note lists files in upload order, and the Agents route receives promptPrefix as a template that the server resolves against the conversation start instead of the browser clock. endpoints.agents.promptMessageTimestamps (default false) states the send time on each user message. Co-authored-by: Lia <lia@librechat.ai>
No file changes beyond LibreChat-AI#16369. Gives LibreChat-AI#17146 and LibreChat-AI#17189 one merge base.
…LibreChat-AI#17147) No file changes beyond LibreChat-AI#16369. Gives LibreChat-AI#17147 and LibreChat-AI#17189 one merge base.
With endpoints.agents.promptCache.stableContext, memory, project, artifacts, skills and the date are sampled at Send; changes reach the model as <context_update> on the user message and the system prompt is rendered from a per-segment anchor. Includes LibreChat-AI#17145.
With endpoints.agents.promptCache.stableContext, MCP servers, actions, agent tools and built-in capabilities are tool groups bound from the segment anchor; a group enabled mid-chat is announced and called through the SDK dispatcher, so tools never change. Changes the dispatcher cannot absorb (model, agent setup, provider tools, schema profile, MCP server hidden by a profile) re-anchor the segment on the sent message. Includes LibreChat-AI#17148.
This was referenced Oct 28, 2026
Contributor
Author
|
Reduced to minimize dependencies and the review size: includes #17148 (closed), +11640 -> +7322 lines against |
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Providers serialize
toolsahead of the system prompt, so switching an MCP server, an action, an agent tool or a built-in capability mid-conversation resets the prompt cache from the first byte. Withendpoints.agents.promptCache.stableContext(#16369) these become tool groups bound from the segment anchor of #17146: a group enabled mid-chat is announced once in the user message and called through the SDK dispatchercall_tool, a group switched off stays bound and is markeddisabled, sotoolsnever changes within a segment. Changes the dispatcher cannot absorb are declared prefix resets that re-anchor the segment on that message. Includes #17148 (built-in capabilities and resets).Related to #17102. Based on #17146. Depends on LibreChat-AI/agents#614 (
toolDispatch); with an SDK without it no tool groups are sampled (tools as loaded, one warning), the rest of the setting keeps working.How it works
toolschanges at the next anchor. Tools found bytool_searchstay deferred and go through the dispatcher.packages/api;MCPManagerkeeps its API.Results
Live, Claude Haiku 5.5, thinking medium, cache TTL 1h, all PRs of #17102 together: G2 (web search and three MCP servers toggled mid-chat and called) 85.1% cached, 0 misses; tools with thinking 87.4%, 0 misses; no API errors.
Changes from the previous version
dev+11640 -288 (old feat: Route Built-in Capabilities Through Tool Groups and Declare Prefix Resets #17148) -> +7322 -230.Type of change
Testing
Enable the setting, switch an MCP server on after the second message and ask for one of its tools: request
toolsstay the same, the announcement is in the user message, the call shows as the real tool in the UI.Automated tests:
client.test.js,ToolService.spec.js,PromptCacheResetNotice.test.tsx,dispatched.test.ts,run-summarization.test.ts,load.spec.ts,promptCache.spec.ts,toolDispatchName.spec.ts,builtins.spec.ts,prepare.spec.ts,tools.spec.ts,turnContext.spec.ts,instructions.spec.ts,definitions.spec.ts,steps.spec.ts,toolDispatch.spec.ts. Guards:run-summarization.test.ts"hands the SDK the tools as upstream without stableContext" (inline snapshot recorded ondev), MCP "keeps the picker order and the catalog order without stableContext". Against the npm SDK only the tests that need the dispatcher fail.tsc --noEmitin packages/api, packages/data-provider, packages/chat, client; eslint and prettier on the touched files.Risk / compatibility
Default off: unchanged. With the setting, MCP servers and their tools of a new conversation are sorted by name.
Checklist