Skip to main content
Glama

vora_update_agent

Update a voice agent incrementally by adding objection responses, updating pricing or context, applying AI-driven improvements from call analytics, refining objectives, or triggering a full recompile. Changes apply immediately on next call.

Instructions

Incrementally improve your voice agent without full recompilation. Use this to:

  • Add new objection responses based on what you're hearing in calls

  • Update pricing or product information

  • Apply AI-recommended improvements from call analytics (apply_learnings)

  • Refine your objective based on what's working

  • Force a full recompile when your website has significantly changed

Changes take effect on the very next call. The apply_learnings option is powerful — Vora analyzes all your past calls and automatically updates objection handling, greetings, and qualification criteria based on what's actually converting.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
agent_idYesThe voice agent to update.
add_objection_responseNoAdd a specific objection-response pair.
update_pricingNoNew pricing information. Replaces the current pricing in the pitch.
update_contextNoAdditional business context. New product features, changed messaging, new competitors, etc.
update_objectiveNoRefined calling objective based on what's working. E.g., switch from 'qualify for enterprise' to 'book demo for SMBs'.
apply_learningsNoAuto-apply Vora's recommended improvements from call analytics. Analyzes all past calls and updates objection handling, greetings, and qualification criteria. Powerful after 10+ calls.
recompileNoFull recompilation from your website URL + all accumulated context. Use when your website has significantly changed. Preserves all call learnings.

Implementation Reference

  • The async handler function that calls Vora API with update parameters and returns a formatted success/error response.
        async (params) => {
          const client = getApiClient();
    
          try {
            const response = await client.patch<UpdateAgentResponse>(
              `/v1/agent-api/agents/${params.agent_id}`,
              {
                add_objection_response: params.add_objection_response,
                update_pricing: params.update_pricing,
                update_context: params.update_context,
                update_objective: params.update_objective,
                apply_learnings: params.apply_learnings,
                recompile: params.recompile,
              }
            );
    
            const lines: string[] = [
              `Agent ${response.agent_id} updated!`,
              `Context Score: ${response.context_score}/100`,
              `Updated: ${response.updated_fields.join(", ")}`,
            ];
    
            if (response.applied_learnings && response.applied_learnings.length > 0) {
              lines.push(`\nApplied Learnings:`);
              response.applied_learnings.forEach((l) => lines.push(`  - ${l}`));
            }
    
            if (response.recompiled) {
              lines.push(`\nFull recompile completed. Website re-crawled and agent rebuilt with all accumulated learnings preserved.`);
            }
    
            lines.push(`\n${response.message}`);
    
            return {
              content: [{ type: "text" as const, text: lines.join("\n") }],
            };
          } catch (error) {
            return {
              content: [
                {
                  type: "text" as const,
                  text: `Update error: ${error instanceof Error ? error.message : String(error)}`,
                },
              ],
              isError: true,
            };
          }
        }
      );
    }
  • Zod schema defining input parameters: agent_id, add_objection_response, update_pricing, update_context, update_objective, apply_learnings, recompile.
      agent_id: z
        .string()
        .describe("The voice agent to update."),
      add_objection_response: z
        .object({
          objection: z
            .string()
            .describe(
              "The objection to handle. E.g., 'We already use Toast' or 'Too expensive'"
            ),
          response: z
            .string()
            .describe(
              "How to respond. E.g., 'Many of our best customers switched from Toast because of our Arabic language support...'"
            ),
        })
        .optional()
        .describe("Add a specific objection-response pair."),
      update_pricing: z
        .string()
        .optional()
        .describe(
          "New pricing information. Replaces the current pricing in the pitch."
        ),
      update_context: z
        .string()
        .optional()
        .describe(
          "Additional business context. New product features, changed messaging, new competitors, etc."
        ),
      update_objective: z
        .string()
        .optional()
        .describe(
          "Refined calling objective based on what's working. E.g., switch from 'qualify for enterprise' to 'book demo for SMBs'."
        ),
      apply_learnings: z
        .boolean()
        .optional()
        .describe(
          "Auto-apply Vora's recommended improvements from call analytics. Analyzes all past calls and updates objection handling, greetings, and qualification criteria. Powerful after 10+ calls."
        ),
      recompile: z
        .boolean()
        .optional()
        .describe(
          "Full recompilation from your website URL + all accumulated context. Use when your website has significantly changed. Preserves all call learnings."
        ),
    },
  • TypeScript interface for the Vora API response (UpdateAgentResponse).
    interface UpdateAgentResponse {
      agent_id: string;
      updated_fields: string[];
      context_score: number;
      applied_learnings?: string[];
      recompiled?: boolean;
      message: string;
    }
  • Registration call that binds registerVoraUpdateAgent to the MCP server.
    registerVoraUpdateAgent(server);
  • The registerVoraUpdateAgent function that registers the tool with the MCP server under the name 'vora_update_agent'.
    export function registerVoraUpdateAgent(server: McpServer): void {
      server.tool(
        "vora_update_agent",
        `Incrementally improve your voice agent without full recompilation. Use this to:
    
    - Add new objection responses based on what you're hearing in calls
    - Update pricing or product information
    - Apply AI-recommended improvements from call analytics (apply_learnings)
    - Refine your objective based on what's working
    - Force a full recompile when your website has significantly changed
    
    Changes take effect on the very next call. The apply_learnings option is powerful — Vora analyzes all your past calls and automatically updates objection handling, greetings, and qualification criteria based on what's actually converting.`,
        {
          agent_id: z
            .string()
            .describe("The voice agent to update."),
          add_objection_response: z
            .object({
              objection: z
                .string()
                .describe(
                  "The objection to handle. E.g., 'We already use Toast' or 'Too expensive'"
                ),
              response: z
                .string()
                .describe(
                  "How to respond. E.g., 'Many of our best customers switched from Toast because of our Arabic language support...'"
                ),
            })
            .optional()
            .describe("Add a specific objection-response pair."),
          update_pricing: z
            .string()
            .optional()
            .describe(
              "New pricing information. Replaces the current pricing in the pitch."
            ),
          update_context: z
            .string()
            .optional()
            .describe(
              "Additional business context. New product features, changed messaging, new competitors, etc."
            ),
          update_objective: z
            .string()
            .optional()
            .describe(
              "Refined calling objective based on what's working. E.g., switch from 'qualify for enterprise' to 'book demo for SMBs'."
            ),
          apply_learnings: z
            .boolean()
            .optional()
            .describe(
              "Auto-apply Vora's recommended improvements from call analytics. Analyzes all past calls and updates objection handling, greetings, and qualification criteria. Powerful after 10+ calls."
            ),
          recompile: z
            .boolean()
            .optional()
            .describe(
              "Full recompilation from your website URL + all accumulated context. Use when your website has significantly changed. Preserves all call learnings."
            ),
        },
        async (params) => {
          const client = getApiClient();
    
          try {
            const response = await client.patch<UpdateAgentResponse>(
              `/v1/agent-api/agents/${params.agent_id}`,
              {
                add_objection_response: params.add_objection_response,
                update_pricing: params.update_pricing,
                update_context: params.update_context,
                update_objective: params.update_objective,
                apply_learnings: params.apply_learnings,
                recompile: params.recompile,
              }
            );
    
            const lines: string[] = [
              `Agent ${response.agent_id} updated!`,
              `Context Score: ${response.context_score}/100`,
              `Updated: ${response.updated_fields.join(", ")}`,
            ];
    
            if (response.applied_learnings && response.applied_learnings.length > 0) {
              lines.push(`\nApplied Learnings:`);
              response.applied_learnings.forEach((l) => lines.push(`  - ${l}`));
            }
    
            if (response.recompiled) {
              lines.push(`\nFull recompile completed. Website re-crawled and agent rebuilt with all accumulated learnings preserved.`);
            }
    
            lines.push(`\n${response.message}`);
    
            return {
              content: [{ type: "text" as const, text: lines.join("\n") }],
            };
          } catch (error) {
            return {
              content: [
                {
                  type: "text" as const,
                  text: `Update error: ${error instanceof Error ? error.message : String(error)}`,
                },
              ],
              isError: true,
            };
          }
        }
      );
    }
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations, so description carries full burden. Discloses that changes take effect on the next call and that apply_learnings analyzes past calls. Lacks details on permissions, error handling, or idempotency.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Well-structured with a clear opening, bullet list of use cases, and a concluding note on effect. Could be slightly more concise, but generally efficient.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the complexity of 7 parameters and no output schema, the description covers the main actions and effects. Explains each parameter's role and the overall behavior, though it could mention parameter combinations.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% with detailed descriptions, but the description adds value through concrete examples (e.g., 'We already use Toast' for objection) and usage scenarios, enhancing meaning beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool is for incrementally improving a voice agent without full recompilation, listing specific use cases. It distinguishes from sibling tools like vora_create_agent (creation) and vora_call (calls) by focusing on updates.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides clear context on when to use (for incremental updates) and when to force a full recompile (website significantly changed). Does not explicitly exclude alternatives but implies usage scenarios.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/stefanstojanovicstefa-creator/vora-voice-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server