Melchizedek

Overview Schema Related Servers Score Discussions

m9k_info

Read-onlyIdempotent

Display memory index metrics including corpus size, search pipeline status, usage statistics, and embedding worker state for persistent conversation storage.

Instructions

Show memory index information: corpus size, search pipeline status, usage metrics, embedding worker state.

Input Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

Implementation Reference

src/tools/memory.ts:172-419 (handler)

The handler function for 'm9k_info' that aggregates statistics about the memory index, search pipeline, usage metrics, and current process status.

async () => {
  const sessions = (
    ctx.db
      .prepare('SELECT COUNT(*) AS cnt FROM conv_sessions WHERE deleted_at IS NULL')
      .get() as {
      cnt: number;
    }
  ).cnt;
  const chunks = (
    ctx.db
      .prepare('SELECT COUNT(*) AS cnt FROM conv_chunks WHERE deleted_at IS NULL')
      .get() as {
      cnt: number;
    }
  ).cnt;
  const projects = (
    ctx.db
      .prepare(
        'SELECT COUNT(DISTINCT project) AS cnt FROM conv_sessions WHERE deleted_at IS NULL',
      )
      .get() as { cnt: number }
  ).cnt;

  const oldestSession = (
    ctx.db
      .prepare('SELECT MIN(started_at) AS ts FROM conv_sessions WHERE deleted_at IS NULL')
      .get() as { ts: string | null }
  ).ts;
  const newestSession = (
    ctx.db
      .prepare('SELECT MAX(started_at) AS ts FROM conv_sessions WHERE deleted_at IS NULL')
      .get() as { ts: string | null }
  ).ts;

  // Build span string from oldest/newest dates
  const formatDate = (iso: string | null): string | null => {
    if (!iso) return null;
    return iso.slice(0, 10); // "2026-02-28T..." → "2026-02-28"
  };
  const oldest = formatDate(oldestSession);
  const newest = formatDate(newestSession);
  const span = oldest && newest ? `${oldest} → ${newest}` : null;

  const eligibleChunks = countEligibleChunks(ctx.db);
  let embeddedChunksText = 0;
  let embeddedChunksCode = 0;
  try {
    embeddedChunksText = countEmbeddedChunks(ctx.db, '_text');
  } catch {
    // vec table may not exist yet
  }
  try {
    embeddedChunksCode = countEmbeddedChunks(ctx.db, '_code');
  } catch {
    // vec table may not exist yet
  }

  const ignoredProjects = getIgnoredProjects(ctx.db).length;
  const orphanedSessions = parseInt(getStat(ctx.db, 'orphaned_sessions') ?? '0', 10);

  const searchCount = parseInt(getStat(ctx.db, 'search_count') ?? '0', 10);
  const hitCount = parseInt(getStat(ctx.db, 'hit_count') ?? '0', 10);
  const tokensServed = parseInt(getStat(ctx.db, 'tokens_served') ?? '0', 10);
  const lastSearchAt = getStat(ctx.db, 'last_search_at');

  // --- Search lines: one readable string per embedder ---
  const embeddingModelIdText = getMeta(ctx.db, 'embedding_model_id_text');
  const embeddingDimensionsText = parseInt(
    getMeta(ctx.db, 'embedding_dimensions_text') ?? '0',
    10,
  );
  const embeddingModelIdCode = getMeta(ctx.db, 'embedding_model_id_code');
  const embeddingDimensionsCode = parseInt(
    getMeta(ctx.db, 'embedding_dimensions_code') ?? '0',
    10,
  );

  const buildSearchLine = (
    modelId: string | null,
    dims: number,
    embedded: number,
    eligible: number,
  ): string => {
    if (!modelId) return 'disabled';
    const pct = eligible > 0 ? Math.round((embedded / eligible) * 100) : 0;
    return `${modelId} (${dims}d) — ${embedded}/${eligible} (${pct}%)`;
  };

  const textLine = buildSearchLine(
    embeddingModelIdText,
    embeddingDimensionsText,
    embeddedChunksText,
    eligibleChunks,
  );
  const codeLine = buildSearchLine(
    embeddingModelIdCode,
    embeddingDimensionsCode,
    embeddedChunksCode,
    eligibleChunks,
  );

  // Reranker line
  const rerankerLine = ctx.searchContext.reranker
    ? `${ctx.searchContext.reranker.backend()} (${ctx.searchContext.reranker.modelId()})`
    : 'none';

  // --- Activity: merge worker + migration into one block ---
  const workerStatus: EmbedJobStatus = ctx.orchestrator?.getStatus() ?? {
    active: false,
    suffix: null,
    embedded: 0,
    total: 0,
    pid: null,
    rssMB: null,
    heapUsedMB: null,
  };

  const hasMigrationText = !!getMeta(ctx.db, 'migration_target_model_text');
  const hasMigrationCode = !!getMeta(ctx.db, 'migration_target_model_code');

  let activity: {
    type: string;
    target: string | null;
    progress: string;
    pid: number | null;
    rssMB: number | null;
    heapUsedMB: number | null;
    stuckSince?: string | null;
  } | null = null;

  const buildProgressString = (embedded: number, total: number): string => {
    const pct = total > 0 ? Math.round((embedded / total) * 100) : 0;
    return `${embedded}/${total} (${pct}%)`;
  };

  if (workerStatus.active) {
    // Check if current worker suffix matches a migration
    const suffix = workerStatus.suffix ?? '_text';
    const isMigration =
      (suffix === '_text' && hasMigrationText) || (suffix === '_code' && hasMigrationCode);

    let embedded = workerStatus.embedded;
    let total = workerStatus.total;

    // Worker may not have reported total yet — fall back to migration meta / eligible chunks
    if (isMigration && total === 0) {
      embedded = parseInt(getMeta(ctx.db, `migration_progress${suffix}`) ?? '0', 10);
      total = eligibleChunks;
    }

    activity = {
      type: isMigration ? 'migration' : 'backfill',
      target: workerStatus.suffix,
      progress: buildProgressString(embedded, total),
      pid: workerStatus.pid,
      rssMB: workerStatus.rssMB,
      heapUsedMB: workerStatus.heapUsedMB,
    };
  } else if (hasMigrationText || hasMigrationCode) {
    // Migration meta exists but no worker — stuck
    const suffix = hasMigrationText ? '_text' : '_code';
    const stuckSince = getMeta(ctx.db, `migration_started_at${suffix}`);
    const migrationProgress = parseInt(
      getMeta(ctx.db, `migration_progress${suffix}`) ?? '0',
      10,
    );
    activity = {
      type: 'stuck-migration',
      target: suffix,
      progress: buildProgressString(migrationProgress, eligibleChunks),
      pid: null,
      rssMB: null,
      heapUsedMB: null,
      stuckSince,
    };
  }

  // --- Status emoji ---
  let status: string;
  if (!ctx.cfg.embeddingsEnabled) {
    status = '\u{1F7E2} bm25-only';
  } else if (workerStatus.active) {
    status = '\u{1F535} embedding';
  } else if (activity?.type === 'stuck-migration') {
    status = '\u{1F7E0} degraded';
  } else {
    status = '\u{1F7E2} healthy';
  }

  // --- Uptime ---
  const uptimeSec = Math.round(process.uptime());
  let uptime: string;
  if (uptimeSec < 60) {
    uptime = `${uptimeSec}s`;
  } else if (uptimeSec < 3600) {
    uptime = `${Math.floor(uptimeSec / 60)}m`;
  } else {
    const h = Math.floor(uptimeSec / 3600);
    const m = Math.floor((uptimeSec % 3600) / 60);
    uptime = m > 0 ? `${h}h${m}m` : `${h}h`;
  }

  // --- Process ---
  const mem = process.memoryUsage();
  const processInfo = {
    pid: process.pid,
    rssMB: Math.round((mem.rss / 1024 / 1024) * 10) / 10,
    heapUsedMB: Math.round((mem.heapUsed / 1024 / 1024) * 10) / 10,
    uptimeSeconds: uptimeSec,
  };

  return {
    content: [
      {
        type: 'text' as const,
        text: JSON.stringify({
          version: ctx.version,
          status,
          mode: ctx.mode,
          logLevel: ctx.cfg.logLevel,
          uptime,
          corpus: {
            sessions,
            chunks,
            projects,
            ignoredProjects,
            orphanedSessions,
            span,
          },
          search: {
            bm25: 'active',
            text: textLine,
            code: codeLine,
            reranker: rerankerLine,
          },
          activity,
          usage: {
            searches: searchCount,
            hits: hitCount,
            tokensServed,
            lastSearch: lastSearchAt,
          },
          process: processInfo,
        }),
      },
    ],
  };
},

src/tools/memory.ts:159-420 (registration)

Registration of the 'm9k_info' tool within the 'registerMemoryTools' function.

server.registerTool(
  'm9k_info',
  {
    description:
      'Show memory index information: corpus size, search pipeline status, usage metrics, embedding worker state.',
    inputSchema: {},
    annotations: {
      readOnlyHint: true,
      destructiveHint: false,
      idempotentHint: true,
      openWorldHint: false,
    },
  },
  async () => {
    const sessions = (
      ctx.db
        .prepare('SELECT COUNT(*) AS cnt FROM conv_sessions WHERE deleted_at IS NULL')
        .get() as {
        cnt: number;
      }
    ).cnt;
    const chunks = (
      ctx.db
        .prepare('SELECT COUNT(*) AS cnt FROM conv_chunks WHERE deleted_at IS NULL')
        .get() as {
        cnt: number;
      }
    ).cnt;
    const projects = (
      ctx.db
        .prepare(
          'SELECT COUNT(DISTINCT project) AS cnt FROM conv_sessions WHERE deleted_at IS NULL',
        )
        .get() as { cnt: number }
    ).cnt;

    const oldestSession = (
      ctx.db
        .prepare('SELECT MIN(started_at) AS ts FROM conv_sessions WHERE deleted_at IS NULL')
        .get() as { ts: string | null }
    ).ts;
    const newestSession = (
      ctx.db
        .prepare('SELECT MAX(started_at) AS ts FROM conv_sessions WHERE deleted_at IS NULL')
        .get() as { ts: string | null }
    ).ts;

    // Build span string from oldest/newest dates
    const formatDate = (iso: string | null): string | null => {
      if (!iso) return null;
      return iso.slice(0, 10); // "2026-02-28T..." → "2026-02-28"
    };
    const oldest = formatDate(oldestSession);
    const newest = formatDate(newestSession);
    const span = oldest && newest ? `${oldest} → ${newest}` : null;

    const eligibleChunks = countEligibleChunks(ctx.db);
    let embeddedChunksText = 0;
    let embeddedChunksCode = 0;
    try {
      embeddedChunksText = countEmbeddedChunks(ctx.db, '_text');
    } catch {
      // vec table may not exist yet
    }
    try {
      embeddedChunksCode = countEmbeddedChunks(ctx.db, '_code');
    } catch {
      // vec table may not exist yet
    }

    const ignoredProjects = getIgnoredProjects(ctx.db).length;
    const orphanedSessions = parseInt(getStat(ctx.db, 'orphaned_sessions') ?? '0', 10);

    const searchCount = parseInt(getStat(ctx.db, 'search_count') ?? '0', 10);
    const hitCount = parseInt(getStat(ctx.db, 'hit_count') ?? '0', 10);
    const tokensServed = parseInt(getStat(ctx.db, 'tokens_served') ?? '0', 10);
    const lastSearchAt = getStat(ctx.db, 'last_search_at');

    // --- Search lines: one readable string per embedder ---
    const embeddingModelIdText = getMeta(ctx.db, 'embedding_model_id_text');
    const embeddingDimensionsText = parseInt(
      getMeta(ctx.db, 'embedding_dimensions_text') ?? '0',
      10,
    );
    const embeddingModelIdCode = getMeta(ctx.db, 'embedding_model_id_code');
    const embeddingDimensionsCode = parseInt(
      getMeta(ctx.db, 'embedding_dimensions_code') ?? '0',
      10,
    );

    const buildSearchLine = (
      modelId: string | null,
      dims: number,
      embedded: number,
      eligible: number,
    ): string => {
      if (!modelId) return 'disabled';
      const pct = eligible > 0 ? Math.round((embedded / eligible) * 100) : 0;
      return `${modelId} (${dims}d) — ${embedded}/${eligible} (${pct}%)`;
    };

    const textLine = buildSearchLine(
      embeddingModelIdText,
      embeddingDimensionsText,
      embeddedChunksText,
      eligibleChunks,
    );
    const codeLine = buildSearchLine(
      embeddingModelIdCode,
      embeddingDimensionsCode,
      embeddedChunksCode,
      eligibleChunks,
    );

    // Reranker line
    const rerankerLine = ctx.searchContext.reranker
      ? `${ctx.searchContext.reranker.backend()} (${ctx.searchContext.reranker.modelId()})`
      : 'none';

    // --- Activity: merge worker + migration into one block ---
    const workerStatus: EmbedJobStatus = ctx.orchestrator?.getStatus() ?? {
      active: false,
      suffix: null,
      embedded: 0,
      total: 0,
      pid: null,
      rssMB: null,
      heapUsedMB: null,
    };

    const hasMigrationText = !!getMeta(ctx.db, 'migration_target_model_text');
    const hasMigrationCode = !!getMeta(ctx.db, 'migration_target_model_code');

    let activity: {
      type: string;
      target: string | null;
      progress: string;
      pid: number | null;
      rssMB: number | null;
      heapUsedMB: number | null;
      stuckSince?: string | null;
    } | null = null;

    const buildProgressString = (embedded: number, total: number): string => {
      const pct = total > 0 ? Math.round((embedded / total) * 100) : 0;
      return `${embedded}/${total} (${pct}%)`;
    };

    if (workerStatus.active) {
      // Check if current worker suffix matches a migration
      const suffix = workerStatus.suffix ?? '_text';
      const isMigration =
        (suffix === '_text' && hasMigrationText) || (suffix === '_code' && hasMigrationCode);

      let embedded = workerStatus.embedded;
      let total = workerStatus.total;

      // Worker may not have reported total yet — fall back to migration meta / eligible chunks
      if (isMigration && total === 0) {
        embedded = parseInt(getMeta(ctx.db, `migration_progress${suffix}`) ?? '0', 10);
        total = eligibleChunks;
      }

      activity = {
        type: isMigration ? 'migration' : 'backfill',
        target: workerStatus.suffix,
        progress: buildProgressString(embedded, total),
        pid: workerStatus.pid,
        rssMB: workerStatus.rssMB,
        heapUsedMB: workerStatus.heapUsedMB,
      };
    } else if (hasMigrationText || hasMigrationCode) {
      // Migration meta exists but no worker — stuck
      const suffix = hasMigrationText ? '_text' : '_code';
      const stuckSince = getMeta(ctx.db, `migration_started_at${suffix}`);
      const migrationProgress = parseInt(
        getMeta(ctx.db, `migration_progress${suffix}`) ?? '0',
        10,
      );
      activity = {
        type: 'stuck-migration',
        target: suffix,
        progress: buildProgressString(migrationProgress, eligibleChunks),
        pid: null,
        rssMB: null,
        heapUsedMB: null,
        stuckSince,
      };
    }

    // --- Status emoji ---
    let status: string;
    if (!ctx.cfg.embeddingsEnabled) {
      status = '\u{1F7E2} bm25-only';
    } else if (workerStatus.active) {
      status = '\u{1F535} embedding';
    } else if (activity?.type === 'stuck-migration') {
      status = '\u{1F7E0} degraded';
    } else {
      status = '\u{1F7E2} healthy';
    }

    // --- Uptime ---
    const uptimeSec = Math.round(process.uptime());
    let uptime: string;
    if (uptimeSec < 60) {
      uptime = `${uptimeSec}s`;
    } else if (uptimeSec < 3600) {
      uptime = `${Math.floor(uptimeSec / 60)}m`;
    } else {
      const h = Math.floor(uptimeSec / 3600);
      const m = Math.floor((uptimeSec % 3600) / 60);
      uptime = m > 0 ? `${h}h${m}m` : `${h}h`;
    }

    // --- Process ---
    const mem = process.memoryUsage();
    const processInfo = {
      pid: process.pid,
      rssMB: Math.round((mem.rss / 1024 / 1024) * 10) / 10,
      heapUsedMB: Math.round((mem.heapUsed / 1024 / 1024) * 10) / 10,
      uptimeSeconds: uptimeSec,
    };

    return {
      content: [
        {
          type: 'text' as const,
          text: JSON.stringify({
            version: ctx.version,
            status,
            mode: ctx.mode,
            logLevel: ctx.cfg.logLevel,
            uptime,
            corpus: {
              sessions,
              chunks,
              projects,
              ignoredProjects,
              orphanedSessions,
              span,
            },
            search: {
              bm25: 'active',
              text: textLine,
              code: codeLine,
              reranker: rerankerLine,
            },
            activity,
            usage: {
              searches: searchCount,
              hits: hitCount,
              tokensServed,
              lastSearch: lastSearchAt,
            },
            process: processInfo,
          }),
        },
      ],
    };
  },
);

Tool Definition Quality

A4.2/5.0

Behavior4/5

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

Annotations declare readOnly/idempotent/safe properties; the description adds valuable content context by listing exactly what metrics are returned (corpus size, pipeline status, usage metrics, worker state), helping the agent understand the informational scope beyond just knowing it's safe to call.

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

Conciseness5/5

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

Single sentence with optimal structure: action ('Show') + resource ('memory index information') + specific details. Every word earns its place; no waste.

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?

Adequate for a simple read-only status tool; the description compensates for missing output schema by listing returned information categories. Would benefit from guidance distinguishing it from m9k_full or other diagnostic tools.

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?

Zero parameters present per schema, triggering baseline score of 4. Description appropriately does not discuss parameters since none exist.

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?

Uses specific verb 'Show' with clear resource 'memory index information', and distinguishes itself from action-oriented siblings (delete, forget, restart, save) by focusing on status/display. The colon-separated list clarifies exactly what information categories are returned.

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?

Specifies what information is returned (implying diagnostic use), but provides no explicit when-to-use guidance or comparison to siblings like m9k_full, m9k_errors, or m9k_context that might return overlapping information.

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

Tool Definition Quality Score (TDQS)
By punkpeye on April 3, 2026.
mcp
The Hackers Who Tracked My Sleep Cycle
By punkpeye on March 26, 2026.
security
Open Source Has a Bot Problem
By punkpeye on March 19, 2026.
open source

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/louis49/melchizedek'

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