Skip to main content
Glama

AroFlo: Get TaskExpenses

aroflo_get_taskexpenses
Read-onlyIdempotent

Retrieve task expense data from AroFlo with flexible filtering, sorting, and joining options to analyze project costs and manage financial tracking.

Instructions

Query the AroFlo TaskExpenses zone (GET). Use pipe-delimited WHERE clauses like "and|field|=|value", ORDER clauses like "field|asc", and JOIN areas like "lineitems". where/order/join accept either a single string or an array. mode: data|verbose|debug|raw (default: data). Set compact=true and optionally select=["field","nested.field"] to reduce payload size. See resource "aroflo://docs/api/" (example: "aroflo://docs/api/quotes") for valid fields/values.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
whereNo
orderNo
joinNo
pageNo
pageSizeNo
autoPaginateNo
maxPagesNo
maxResultsNo
maxItemsTotalNo
validateWhereNo
modeNo
verboseNo
debugNo
rawNo
compactNo
selectNo
maxItemsNo
extraNo

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Implementation Reference

  • The tool `aroflo_get_taskexpenses` is dynamically registered by iterating over the `AROFLO_ZONES` constant. This generic handler function fetches data from the AroFlo API for the specified zone, including support for pagination, filtering, and output modes.
    server.registerTool(
      toolName,
      {
        title: `AroFlo: Get ${zone}`,
        description:
          `Query the AroFlo ${zone} zone (GET). ` +
          `Use pipe-delimited WHERE clauses like "and|field|=|value", ORDER clauses like "field|asc", and JOIN areas like "lineitems". ` +
          `where/order/join accept either a single string or an array. ` +
          `mode: data|verbose|debug|raw (default: data). ` +
          `Set compact=true and optionally select=[\"field\",\"nested.field\"] to reduce payload size. ` +
          `See resource "aroflo://docs/api/<slug>" (example: "aroflo://docs/api/quotes") for valid fields/values.`,
        inputSchema,
        // MCP SDK expects output schemas to be object schemas (or raw object shapes).
        // `z.any()` causes output validation to crash under the current SDK.
        outputSchema: z.object({}).passthrough(),
        annotations: {
          readOnlyHint: true,
          idempotentHint: true,
          openWorldHint: true
        }
      },
      async (args) => {
        const mode = resolveOutputMode(args);
        const envelopeRequested =
          typeof args.mode === 'string' || Boolean(args.raw) || Boolean(args.verbose);
        try {
          const where = normalizeWhereParam(args.where);
          const order = normalizeOrderParam(args.order);
          const join = normalizeJoinParam(args.join);
    
          if (args.validateWhere !== false) {
            await validateWhereOrThrow({ zone, where });
          }
    
          const autoPaginate = Boolean(args.autoPaginate);
          const startPage = args.page ?? 1;
          const pageSize = args.pageSize ?? (autoPaginate ? 200 : undefined);
          const maxResultsRaw =
            typeof args.maxItemsTotal === 'number' ? args.maxItemsTotal : args.maxResults;
          const maxResults =
            typeof args.maxResults === 'number' && typeof args.maxItemsTotal === 'number'
              ? Math.min(args.maxResults, args.maxItemsTotal)
              : maxResultsRaw;
          const maxPages = args.maxPages ?? (autoPaginate ? 25 : undefined);
    
          const debugInfo: Record<string, unknown> | undefined = args.debug
            ? {
                zone,
                normalized: {
                  where,
                  order,
                  join,
                  page: startPage,
                  pageSize,
                  extra: args.extra
                }
              }
            : undefined;
    
          let response = await client.get(zone, {
            where,
            order,
            join,
            page: startPage,
            pageSize,
            extra: args.extra
          });
    
          let pagesFetched = 1;
          let truncated = false;
          let truncatedReason: string | undefined;
          let nextPage: number | undefined;
    
          if (autoPaginate) {
            let currentPage = startPage;
            let lastPageCount = extractZoneItems(zone, response.data).items.length;
            while (true) {
              const total = extractZoneItems(zone, response.data).items.length;
    
              if (typeof maxResults === 'number' && total >= maxResults) {
                truncated = true;
                truncatedReason = 'maxResults';
                nextPage = currentPage + 1;
                break;
              }
    
              if (typeof maxPages === 'number' && pagesFetched >= maxPages) {
                truncated = true;
                truncatedReason = 'maxPages';
                nextPage = currentPage + 1;
                break;
              }
    
              if (typeof pageSize === 'number' && lastPageCount < pageSize) {
                break;
              }
    
              currentPage += 1;
              const next = await client.get(zone, {
                where,
                order,
                join,
                page: currentPage,
                pageSize,
                extra: args.extra
              });
    
              // If the next page contributes nothing, stop.
              const nextCount = extractZoneItems(zone, next.data).items.length;
              if (nextCount === 0) {
                break;
              }
    
              response = {
                ...response,
                data: mergeZoneResponseData(response.data, next.data).merged
              };
              pagesFetched += 1;
              lastPageCount = nextCount;
    
              // Stop if the last page was short.
              if (typeof pageSize === 'number' && nextCount < pageSize) {
                break;
              }
            }
          }
    
          if (typeof maxResults === 'number') {
            const total = extractZoneItems(zone, response.data).items.length;
            if (total > maxResults) {
              const { truncated: newData } = truncateZoneArrays(response.data, maxResults);
              response = { ...response, data: newData };
              truncated = true;
              truncatedReason = truncatedReason ?? 'maxResults';
            }
          }
    
          let compactApplied = false;
          let effectiveResponse = response;
          const defaultSelect =
            zone === 'Tasks' && args.compact && (!args.select || args.select.length === 0)
              ? [
                  'taskid',
                  'jobnumber',
                  'status',
                  'taskname',
                  'daterequested',
                  'createdutc',
                  'clientid',
                  'org.orgid',
                  'org.orgname',
                  'projectid',
                  'stageid',
                  'project.projectid',
                  'project.projectname',
                  'tasktotals.totalhrs'
                ]
              : undefined;
          const select = args.select ?? defaultSelect;
    
          if (args.compact || (select && select.length > 0) || args.maxItems) {
            compactApplied = true;
            const compactedData = compactZoneResponseData(response.data, {
              select,
              maxItems: args.maxItems
            });
            effectiveResponse = { ...response, data: compactedData };
          }
    
          // Backward compatible default: return the full AroFlo client response, optionally
          // annotated with pagination/debug metadata. The new minimal envelope is opt-in via
          // args.mode / args.verbose / args.raw.
          if (!envelopeRequested) {
            let finalData: unknown = effectiveResponse.data;
            if (autoPaginate || truncated) {
              finalData = withZoneResponseMeta(finalData, {
                pagesFetched,
                truncated,
                truncatedReason,
                nextPage
              });
            }
            if (debugInfo) {
              finalData = withDebug(finalData, debugInfo);
            }
            return successToolResult({ ...effectiveResponse, data: finalData });
          }
    
          const out = buildZoneDataEnvelope({
            zone,
            response: effectiveResponse,
            page: startPage,
            pageSize,
            mode,
            mcp:
              autoPaginate || truncated
                ? {
                    autoPaginate,
                    pagesFetched,
                    truncated,
                    truncatedReason,
                    nextPage
                  }
                : undefined,
            debug: debugInfo,
            compactApplied,
            select,
            maxItems: args.maxItems
          });
    
          return successToolResult(out);
        } catch (error) {
          return errorToolResult(error, { mode, debug: { zone } });
        }
      }
    );
Behavior4/5

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

Annotations declare readOnlyHint/idempotentHint (safe to call); description adds valuable behavioral context including the GET method, four response modes (data|verbose|debug|raw), and payload reduction options (compact/select). No contradictions with annotations.

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?

Five information-dense sentences with zero filler. Front-loaded with the core action (Query), followed by syntax patterns, response modes, optimization flags, and documentation reference. Slightly fragmented sentence structure ('mode: ...') but highly efficient.

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

Completeness3/5

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

With an output schema present, the description appropriately omits return value details. However, given 18 undocumented input parameters and complex query patterns, the description should mention pagination behavior or note that pagination parameters follow standard patterns.

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

Parameters3/5

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

Schema has 0% description coverage for 18 parameters, requiring heavy description compensation. Description thoroughly documents complex query syntax (where/order/join), mode options, and select syntax, but omits 12 parameters including pagination controls (page, pageSize, autoPaginate, maxPages), boolean flags (verbose, debug, raw), and the 'extra' object.

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?

Specific verb 'Query' plus specific resource 'AroFlo TaskExpenses zone' clearly identifies the operation. The '(GET)' annotation and mention of the specific zone distinguishes it from the many sibling getter tools (e.g., get_tasks, get_tasklabours) and the mutation tool (create_or_update_record).

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

Usage Guidelines3/5

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

Provides detailed syntax guidance on how to construct queries (pipe-delimited WHERE, ORDER, JOIN patterns) and references external documentation resources for valid fields. However, lacks explicit when-to-use guidance versus alternative query tools like the generic aroflo_query_zone or specific task-related getters.

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/0x1NotMe/aroflo-mcp'

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