[WIP] Update log analysis tools for MCP server based on feedback #10

Copilot · 2026-01-25T01:23:07Z

Thanks for the feedback on #6. I've created this new PR, which merges into #6, to address your comment. I will work on the changes and keep this PR's description up to date as I make progress.

Original PR: #6
Triggering review: #6 (comment)

@copilot open a new pull request to apply changes based on this feedback

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

Add MCP server support to PM2 for process management through MCP-compatible clients. Features: - New pm2-mcp binary that exposes PM2 process management via MCP - 12 MCP tools for process lifecycle, logging, and monitoring: - pm2_list_processes, pm2_describe_process - pm2_start_process, pm2_restart_process, pm2_reload_process - pm2_stop_process, pm2_delete_process - pm2_flush_logs, pm2_reload_logs, pm2_tail_logs - pm2_dump, pm2_kill_daemon - 2 MCP resources for real-time process information: - pm2://processes (list) - pm2://process/{id} (detail) - Automatic sandbox environment detection and adaptation - Support for stdio and HTTP (Streamable) transports - Client notifications for sandbox status and recommendations - Compatible with Claude Code, Codex, and other MCP clients Implementation: - New lib/mcp/server.js with full MCP server implementation - Uses @modelcontextprotocol/sdk for MCP protocol - Sandbox detection checks home directory writability and environment - Auto-selects writable PM2_HOME in sandboxed environments - No-daemon mode by default for MCP client compatibility - Comprehensive environment variable configuration Documentation: - README with MCP server quickstart and setup commands - Environment variables table (PM2_MCP_*, PM2_HOME, etc.) - Sandbox detection explanation - Tool and resource documentation - Justfile recipes for easy registration with MCP clients Related: - Enables pkgx packaging: pkgxdev/pantry#11219 - Development fork: https://github.com/PromptExecution/pm2-mcp - MCP Specification: https://modelcontextprotocol.io/ Co-authored-by: Claude <noreply@anthropic.com>

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Brian Horakh <35611074+elasticdotventures@users.noreply.github.com>

Signed-off-by: Brian Horakh <35611074+elasticdotventures@users.noreply.github.com>

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>

* Initial plan * Add explanation of Codex CLI tool in README Co-authored-by: elasticdotventures <35611074+elasticdotventures@users.noreply.github.com> --------- Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> Co-authored-by: elasticdotventures <35611074+elasticdotventures@users.noreply.github.com>

Copilot

Pull request overview

This PR extends the MCP server with semantic process state and log-analysis capabilities, adds HTTP hardening timeouts, and updates documentation for platform support and Codex usage. It builds on PR #6 by enriching process descriptions, adding new tools/resources for log inspection and safe descriptions, and tightening server behavior.

Changes:

Add connection coalescing for ensureConnected, HTTP server timeouts, and a suite of helpers for log-tail reading, log pattern analysis, and semantic state inference.
Enrich existing tools/resources (pm2_list_processes, pm2_describe_process, and process resources) with a semantic_state field and introduce new tools pm2_analyze_logs and pm2_describe_process_safe for structured log analysis and privacy-safe process descriptions.
Update README platform support wording and add a brief section describing how to use the MCP server with OpenAI Codex.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 6 comments.

File	Description
`lib/mcp/server.js`	Adds connectionPromise-based connection handling, HTTP timeouts (with currently undefined timeout constants), environment redaction helpers, log pattern/progress analysis, semantic-state heuristics, new MCP tools for log analysis and safe describe, and semantic enrichment of process tools/resources.
`README.md`	Clarifies supported Node.js/Bun versions and briefly documents using the MCP server with OpenAI’s Codex CLI.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-01-25T01:38:41Z

lib/mcp/server.js

+    wrapTool('pm2_analyze_logs', async ({ process, timeframe_minutes, lines }) => {
+      try {
+        await ensureConnected();
+        const description = await pm2Describe(process);
+        if (!description || description.length === 0) {
+          throw new Error(`No process found for "${process}"`);
+        }
+        const env = description[0].pm2_env || {};
+        const { lines: logLines, logPath } = await readRecentLogLines(env, lines);
+        const logAnalysis = analyzeLogPatterns(logLines);
+        const semantic = buildSemanticStateFromHeuristics({
+          env,
+          monit: description[0].monit,
+          logAnalysis,
+          logInfo: { lastModified: null }
+        });
+
+        let suggested_action = 'none';
+        if (logAnalysis.errorsFound.length > 0) suggested_action = 'investigate';
+        else if (logAnalysis.topPattern?.semanticStatus === 'downloading') suggested_action = 'wait_for_completion';
+        else if (logAnalysis.topPattern?.semanticStatus === 'degraded') suggested_action = 'investigate';
+
+        const payload = {
+          process,
+          timeframe_minutes,
+          analysis: {


The timeframe_minutes input parameter is accepted and returned in the pm2_analyze_logs tool, but it is not actually used to limit or filter the logs read from disk, which may confuse clients expecting time-bounded analysis. Either apply this parameter when selecting log lines (for example, by comparing mtime or parsing timestamps) or clarify/remove it from the schema and response.

Copilot · 2026-01-25T01:38:41Z

lib/mcp/server.js

+function sanitizeDescriptionEntry(entry, options = {}) {
+  const cloned = JSON.parse(JSON.stringify(entry || {}));
+  const env = cloned.pm2_env || {};
+
+  if (!options.includeEnvironment) {
+    delete env.env;
+    delete env.env_pm2;
+  } else {
+    env.env = filterEnvironment(env.env, options.environmentFilter, options.redactSecrets);
+    env.env_pm2 = filterEnvironment(env.env_pm2, options.environmentFilter, options.redactSecrets);
+  }
+
+  cloned.pm2_env = env;
+  return cloned;


sanitizeDescriptionEntry clones the entire PM2 description via JSON.parse(JSON.stringify(entry || {})), which will drop non-JSON-safe data such as Dates, Buffers, or functions that might exist in pm2_env and could be useful for debugging. If you need a defensive copy, consider a more targeted shallow/deep clone of the properties you care about so structure is preserved without unintentionally altering types.

Copilot · 2026-01-25T01:38:41Z

lib/mcp/server.js

+  server.registerTool(
+    'pm2_describe_process_safe',
+    {
+      title: 'Describe a PM2 process (privacy-safe)',
+      description: 'Describe a process with optional environment filtering and secret redaction.',
+      inputSchema: describeSafeSchema
+    },
+    wrapTool('pm2_describe_process_safe', async ({ process, include_environment, environment_filter, redact_secrets }) => {
+      try {
+        await ensureConnected();
+        const description = await pm2Describe(process);
+        if (!description || description.length === 0) {
+          throw new Error(`No process found for "${process}"`);
+        }
+
+        const sanitized = await Promise.all(
+          description.map(async item => ({
+            ...sanitizeDescriptionEntry(item, {
+              includeEnvironment: include_environment,
+              environmentFilter: environment_filter,
+              redactSecrets: redact_secrets
+            }),
+            semantic_state: await buildSemanticState(item)
+          }))
+        );
+
+        return {
+          content: textContent(sanitized),
+          structuredContent: { description: sanitized }
+        };


pm2_describe_process_safe introduces new behavior around environment filtering, secret redaction, and inclusion of semantic_state, but there are no tests covering this tool in the existing MCP server test suite. To avoid regressions and to validate that redaction and filtering work as intended (including default vs. overridden options), consider adding targeted tests that exercise this tool and inspect the returned description objects.

Copilot · 2026-01-25T01:38:41Z

lib/mcp/server.js

+async function buildSemanticState(procLike) {
+  const env = procLike.pm2_env || {};
+  const monit = procLike.monit || procLike.pm2_env?.monit || {};
+  const logInfo = await readRecentLogLines(env, 200);
+  const logAnalysis = analyzeLogPatterns(logInfo.lines);
+  return buildSemanticStateFromHeuristics({ env, monit, logAnalysis, logInfo });
+}
+
+async function enrichProcess(proc) {
+  const base = formatProcess(proc);
+  base.semantic_state = await buildSemanticState(proc);
+  return base;
+}


The new semantic state helpers (buildSemanticStateFromHeuristics / buildSemanticState / enrichProcess) add non-trivial heuristics for inferring status and progress from logs and monit data, but there are currently no tests exercising this behavior or asserting the semantic_state shape on tools/resources. Given that this file already has MCP integration tests, it would be valuable to add tests that cover at least a few representative scenarios (e.g., online vs degraded vs stuck, and presence of progress indicators) to prevent regressions.

Copilot · 2026-01-25T01:38:41Z

lib/mcp/server.js

+    // Set timeouts to protect against slow HTTP attacks
+    httpServer.timeout = HTTP_SERVER_TIMEOUT_MS;
+    httpServer.headersTimeout = HTTP_HEADERS_TIMEOUT_MS;
+    httpServer.requestTimeout = HTTP_REQUEST_TIMEOUT_MS;


The new HTTP timeout properties reference HTTP_SERVER_TIMEOUT_MS, HTTP_HEADERS_TIMEOUT_MS, and HTTP_REQUEST_TIMEOUT_MS, but these constants are not defined in this file or elsewhere in the repo, which will throw a ReferenceError when createTransport runs for HTTP. These values should either be defined (e.g., with sensible defaults or environment overrides) or the assignments guarded/removed.

Copilot · 2026-01-25T01:38:42Z

lib/mcp/server.js

+  server.registerTool(
+    'pm2_analyze_logs',
+    {
+      title: 'Analyze PM2 logs',
+      description: 'Parse recent logs for activity, patterns, and errors.',
+      inputSchema: analyzeLogsSchema
+    },
+    wrapTool('pm2_analyze_logs', async ({ process, timeframe_minutes, lines }) => {
+      try {
+        await ensureConnected();
+        const description = await pm2Describe(process);
+        if (!description || description.length === 0) {
+          throw new Error(`No process found for "${process}"`);
+        }
+        const env = description[0].pm2_env || {};
+        const { lines: logLines, logPath } = await readRecentLogLines(env, lines);
+        const logAnalysis = analyzeLogPatterns(logLines);
+        const semantic = buildSemanticStateFromHeuristics({
+          env,
+          monit: description[0].monit,
+          logAnalysis,
+          logInfo: { lastModified: null }
+        });
+
+        let suggested_action = 'none';
+        if (logAnalysis.errorsFound.length > 0) suggested_action = 'investigate';
+        else if (logAnalysis.topPattern?.semanticStatus === 'downloading') suggested_action = 'wait_for_completion';
+        else if (logAnalysis.topPattern?.semanticStatus === 'degraded') suggested_action = 'investigate';
+
+        const payload = {
+          process,
+          timeframe_minutes,
+          analysis: {
+            current_activity: logAnalysis.topPattern?.pattern || semantic.status,
+            detected_patterns: logAnalysis.detectedPatterns,
+            errors_found: logAnalysis.errorsFound,
+            warnings_found: logAnalysis.warningsFound,
+            progress_indicators: logAnalysis.progressIndicators,
+            anomalies: [],
+            suggested_action
+          },
+          meta: {
+            log_path: logPath,
+            semantic_state: semantic
+          }
+        };
+
+        return {
+          content: textContent(payload),
+          structuredContent: payload
+        };


The pm2_analyze_logs tool is newly introduced with custom log pattern and progress analysis, but there are no tests in test/mcp/server.mocha.js (or elsewhere under test/) that call this tool or validate its structured response shape. Adding at least a basic happy-path test that starts a process, invokes pm2_analyze_logs, and asserts key fields (patterns, errors/warnings arrays, semantic_state, etc.) would improve confidence and catch future changes to the analysis heuristics.

elasticdotventures and others added 12 commits December 3, 2025 12:12

Add semantic state and log analysis tools to MCP server

e5be71b

Update README.md

04d43b1

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Update lib/mcp/server.js

8bcb96e

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Update lib/mcp/server.js

0fc5f62

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Update lib/mcp/server.js

53fed7e

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Brian Horakh <35611074+elasticdotventures@users.noreply.github.com>

Update lib/mcp/server.js

7cd4e31

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Brian Horakh <35611074+elasticdotventures@users.noreply.github.com>

Update lib/mcp/server.js

4dc78bf

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Brian Horakh <35611074+elasticdotventures@users.noreply.github.com>

Merge branch 'master' into feature/mcp-semantic-state

7abf249

Signed-off-by: Brian Horakh <35611074+elasticdotventures@users.noreply.github.com>

Initial plan (#8)

cd8eb68

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>

Initial plan

a419334

Copilot AI assigned Copilot and elasticdotventures Jan 25, 2026

Copilot AI mentioned this pull request Jan 25, 2026

Add semantic state and log analysis tools to MCP server #6

Merged

Base automatically changed from feature/mcp-semantic-state to master January 25, 2026 01:23

elasticdotventures approved these changes Jan 25, 2026

View reviewed changes

elasticdotventures marked this pull request as ready for review January 25, 2026 01:23

Copilot AI review requested due to automatic review settings January 25, 2026 01:23

elasticdotventures merged commit 3518cdd into master Jan 25, 2026
1 check failed

elasticdotventures deleted the copilot/sub-pr-6-another-one branch January 25, 2026 01:24

Copilot AI requested a review from elasticdotventures January 25, 2026 01:24

Copilot stopped work on behalf of elasticdotventures due to an error January 25, 2026 01:24
The session was cancelled by the user.

Copilot started reviewing on behalf of elasticdotventures January 25, 2026 01:24 View session

Copilot AI reviewed Jan 25, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[WIP] Update log analysis tools for MCP server based on feedback #10

[WIP] Update log analysis tools for MCP server based on feedback #10

Copilot AI commented Jan 25, 2026

Uh oh!

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Jan 25, 2026

Uh oh!

Copilot AI Jan 25, 2026

Uh oh!

Copilot AI Jan 25, 2026

Uh oh!

Copilot AI Jan 25, 2026

Uh oh!

Copilot AI Jan 25, 2026

Uh oh!

Copilot AI Jan 25, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

[WIP] Update log analysis tools for MCP server based on feedback #10

[WIP] Update log analysis tools for MCP server based on feedback #10

Conversation

Copilot AI commented Jan 25, 2026

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Jan 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 25, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants