Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@SentinelMCP Diagnose why nginx service is down on production server"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
SentinelMCP is an enterprise-grade Model Context Protocol (MCP) server written in TypeScript using the NitroStack framework. It acts as an autonomous execution and monitoring layer for LLM-driven agents to detect, diagnose, repair, verify, and prevent software and hardware failures on local and remote systems.
SentinelMCP is submitted under the Open Innovation track. It tackles a real-world problem that cuts across every industry — unplanned system downtime and manual IT-ops firefighting — by giving LLM-driven agents a safe, autonomous execution layer to detect, diagnose, repair, verify, and prevent software and hardware failures on local and remote systems, with built-in safety checks (OSV.dev CVE database) and automatic rollbacks to protect continuity regardless of domain.
Related MCP server: Triage MCP Server
State Lock Mutex: Enforces single-repair execution globally. Prevents concurrent file updates or conflicting service restarts.
Local Log Processor & Deduplicator: Summarizes system/application logs locally before transferring them to the LLM, reducing token consumption.
Safe Command Executor: Blocks dangerous arbitrary terminal command executions by utilizing a strict whitelist and running commands via execFile (bypassing shell evaluation).
Transparent Safety Engine: Automatically creates checksummed backups of configuration files before modifying them, supporting Automated Rollback if post-repair verification checks fail.
Infinite Loop Breaker: Logs repair history per diagnostic fingerprint, dropping confidence scores or blocking strategies that have repeatedly failed.
Hardware Diagnosis & Cost Lookup: Distinguishes hardware from software issues. Prevents software modifications on bad hardware, outputs recommendations, and looks up estimated replacement/repair costs (in INR and USD).
Remote Targets Support (targetHost): Allows tools to target a specified remote server over secure SSH connections, executing remote diagnostics and configs while maintaining local safety backups.
OSV.dev API Vulnerability Integration: Live API queries to check for package CVEs during root cause analysis.
sentinel-mcp/
├── src/
│ ├── index.ts # Application Entry Point & Bootstrap
│ ├── app.module.ts # Root Module Configuration
│ ├── config/
│ │ ├── env.config.ts # Runtime configuration variables
│ │ └── hardware-costs.json # Hardware component replacement estimates
│ ├── interfaces/ # Typings for Incidents, Backups, and Snapshots
│ ├── modules/
│ │ └── sentinel/
│ │ └── sentinel.module.ts # Sentinel Module Registry
│ ├── prompts/
│ │ └── sentinel.prompts.ts # Reusable prompt templates (diagnose-and-repair, postmortem)
│ ├── resources/
│ │ └── sentinel.resources.ts # Exposes incident logs and prevention rules resources
│ ├── services/ # Shared Dependency Injected Services
│ │ ├── lock.service.ts # Concurrency Mutex
│ │ ├── safety.service.ts # Backup/Rollback Manager
│ │ ├── log.service.ts # Local Log Aggregator & Deduplicator
│ │ └── memory.service.ts # JSON File database storing incident states
│ ├── tools/ # Decorated MCP Tool Controllers
│ │ ├── detection/ # scan_system_logs, get_service_states, get_system_metrics, check_network_status
│ │ ├── diagnosis/ # analyze_root_cause (integrated with OSV.dev API)
│ │ ├── correction/ # execute_service_repair, execute_config_patch, execute_package_repair
│ │ ├── verification/ # verify_repair_state (with automated rollback)
│ │ ├── memory/ # retrieve_similar_incidents, get_incident_history, check_repair_strategy_limit
│ │ └── prevention/ # register_prevention_rule, get_prevention_rules
│ └── utils/
│ ├── command-executor.ts # Whitelisted execFile commander (supports SSH targetHost)
│ └── file-helper.ts # Backup and file operations helper (supports remote files)scan_system_logs: Summarizes systemd errors and warning logs. Supports targetHost.
get_service_states: Checks if specified systemd services are running, stopped, or disabled. Supports targetHost.
get_system_metrics: Retrieves CPU, RAM, disk partitions, sensor temperatures, and S.M.A.R.T indicators. Supports targetHost.
check_network_status: Diagnoses IP configurations, listening TCP sockets (ss), DNS resolution, and gateway pings. Supports targetHost.
analyze_root_cause: Synthesizes evidence to rank repair hypotheses or flag hardware failure indicators. Queries OSV.dev API if a packageName is supplied to discover known CVE vulnerabilities.
execute_service_repair: Restarts, starts, or enables a systemd service under the global lock. Supports targetHost.
execute_config_patch: Safely patches config files after making pre-repair backups. Supports targetHost.
execute_package_repair: Installs or reinstalls missing system packages or npm packages. Supports targetHost.
verify_repair_state: Tests system parameters. If checks fail, automatically restores backup files on the target to return the system to its pre-repair state. Supports targetHost.
retrieve_similar_incidents: Looks up past incidents matching the current error profile.
get_incident_history: Lists all logged incidents.
check_repair_strategy_limit: Blocks repair strategies that have failed 2+ times.
register_prevention_rule: Logs recommended hardiness configuration guidelines.
get_prevention_rules: Lists registered rules.
incident://history: Read-only JSON list of all past diagnostic and self-healing incidents, sorted by recency.
prevention://rules: Read-only JSON list of all registered hardening rules.
diagnose-and-repair: Guides the AI client step-by-step through the detection → analysis → correction → verification flow.
incident-postmortem: Formulates a post-mortem summary request for any resolved incident by its UUID.
Install Dependencies:
npm installConfiguration:
Copy .env.example to .env and set environment variables.
Set OAUTH_REQUIRED=false for local developer STDIO executions.
Set SENTINEL_TARGET_HOST and SENTINEL_TARGET_MODE to target remote machines via SSH.
Build the Server:
npm run buildRun Dev / Dev Client:
npm run devStart Production Server:
npm startFollow this sequence of tool calls in an MCP client (such as Claude Desktop or NitroStudio) to demonstrate SentinelMCP's end-to-end self-healing and automated rollback feature:
Check System Health (Detection):
Call get_service_states with:
{
"services": ["nginx"],
"targetHost": "localhost"
}Result: Shows that the nginx service is inactive or down.
Retrieve Log Patterns (Detection):
Call scan_system_logs with:
{
"service": "nginx",
"maxLines": 50,
"targetHost": "localhost"
}Result: Aggregates recent nginx logs (e.g., "Address already in use").
Diagnose and Scan Vulnerabilities (Diagnosis):
Call analyze_root_cause with:
{
"evidenceFingerprint": "nginx-port-80-conflict",
"severity": "high",
"logs": ["nginx: Address already in use", "nginx: failed to bind to port 80"],
"packageName": "nginx"
}Result: Returns a ranked list of repair hypotheses and queries the OSV.dev API to show any known vulnerabilities for nginx. Creates a new incident in the local database.
Apply Patch (Correction):
Call execute_config_patch with:
{
"filePath": "/etc/nginx/nginx.conf",
"patchContent": "events {} http { server { listen 8080; } }",
"incidentId": "<UUID_FROM_PREVIOUS_STEP>",
"dryRun": false
}Result: Acquires the global repair lock, creates a checksummed backup of the config locally, and writes the new configuration.
Verify State & Demonstrate Rollback (Verification):
To demonstrate safety, request verification of a port that will fail (e.g., port 80):
Call verify_repair_state with:
{
"incidentId": "<UUID_FROM_PREVIOUS_STEP>",
"criteria": {
"serviceName": "nginx",
"listeningPorts": [80]
}
}Result: SentinelMCP detects that the verification checks failed. It immediately triggers a rollback, restoring /etc/nginx/nginx.conf to its original checksummed state, updates the incident database status to failed, releases the global state lock, and notifies the caller.
Perform root cause analysis on system evidence. Outputs confidence scores, severity, and ranked repair hypotheses. Handles hardware faults by recommending replacements and estimating costs. Integrates with OSV.dev CVE database for vulnerability lookup.
| Name | Required | Description | Default |
|---|---|---|---|
| logs | Yes | List of log signatures or entries associated with the failure | |
| metrics | No | Key-value system metrics (e.g. memory usage, SMART message) | |
| severity | Yes | Pre-assessed severity level | |
| packageName | No | Optional package name to lookup vulnerabilities in OSV.dev (e.g., "axios" or "openssl") | |
| packageVersion | No | Optional package version to query OSV.dev database | |
| evidenceFingerprint | Yes | Unique hash representing this specific issue profile |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries full burden. It mentions outputs (confidence scores, severity, ranked hypotheses) and specific behaviors (handles hardware faults, integrates with OSV.dev). However, it does not disclose whether the tool modifies system state, requires authentication, or has rate limits. Transparency is adequate but not comprehensive.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is three sentences, each earning its place: first defines core purpose and outputs, second adds hardware fault handling, third mentions CVE integration. It is concise, front-loaded, and free of fluff. Perfectly sized for quick agent comprehension.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool has 6 parameters, nested objects, and no output schema, the description should elaborate on the return format and integration details (e.g., how OSV.dev lookup works—are both packageName and packageVersion required?). It only vaguely mentions outputs, leaving the agent to infer response structure. Gaps in output specification and integration usage make it incomplete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the baseline is 3. The description adds context by linking packageName and packageVersion to OSV.dev integration and metrics to hardware faults, but does not substantially enhance understanding beyond the schema. It does not explain parameter format or constraints beyond what the schema already provides.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool performs root cause analysis on system evidence, produces confidence scores, severity, and ranked repair hypotheses. It distinguishes from siblings like scan_system_logs and get_system_metrics by mentioning hardware fault handling and OSV.dev integration, making its purpose specific and differentiated.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage when system evidence is available, but it does not explicitly state when to use this tool versus alternatives like scan_system_logs. There is no guidance on prerequisites or when not to use. Usage context is inferred but not directly communicated.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Ensure that the same repair strategy is not retried endlessly. If it has failed multiple times for this failure fingerprint, it will be marked as BLOCKED.
| Name | Required | Description | Default |
|---|---|---|---|
| fingerprint | Yes | The diagnostic fingerprint of the error under repair | |
| strategyName | Yes | The name of the repair strategy to validate |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It discloses that the tool checks failure counts and may mark as BLOCKED, but does not specify the threshold for 'multiple times' or what the BLOCKED state means, leaving behavioral ambiguity.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two concise sentences with no redundant information. Every word adds value, and the key behavior is front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple tool with two parameters and no output schema, the description captures the core purpose and basic behavior. However, it lacks details on the failure threshold and consequences of BLOCKED status, which slightly reduces completeness.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema documentation covers both parameters (fingerprint and strategyName) with clear descriptions (100% coverage). The description adds no additional parameter meaning beyond what the schema already provides, meeting the baseline.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description clearly states the tool ensures a repair strategy is not retried endlessly and marks it as BLOCKED after multiple failures. It uses specific verbs and resources, and distinguishes from sibling tools which are about system capabilities or metrics.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for preventing endless retries but provides no explicit guidance on when to use, when not to use, or alternatives. Sibling tools are unrelated, so no direct comparison is available.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Query the operational capabilities and supported monitoring features on this host platform (Windows or Linux).
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden but only states it's a query operation, which implies read-only. It does not disclose output format, performance implications, or any constraints. This is minimally adequate.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single concise sentence with no redundant or unnecessary words. It front-loads the action and resource efficiently.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the simplicity (no parameters, no output schema), the description is somewhat incomplete—it does not specify what the returned capabilities look like or if any platform-specific differences exist. More detail would improve agent understanding.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
There are zero parameters, so the baseline is 4. The description adds no parameter-specific information, which is acceptable as the schema is fully covered.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb 'Query' and identifies the resource as 'operational capabilities and supported monitoring features' on the host platform, which clearly distinguishes it from sibling tools that focus on logs, metrics, services, or repair strategies.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context ('on this host platform') but does not explicitly state when to avoid using this tool or mention alternatives. Its purpose is distinct enough from siblings to imply appropriate usage.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Fetch the active status, configuration, and state of system services. Supports Windows and Linux services.
| Name | Required | Description | Default |
|---|---|---|---|
| services | Yes | List of service names to query (e.g., ["nginx", "ssh", "postgresql"]) | |
| targetHost | No | Optional remote target host (e.g., user@hostname) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries full burden. It indicates a read operation ('Fetch') and cross-platform support, but lacks details on permissions, error handling, or any side effects. Adequate but not thorough.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is extremely concise: two sentences that clearly convey the purpose and scope. Every word adds value, and the key information is front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The description covers the tool's core function and cross-platform nature. However, without an output schema, mentioning the return format would improve completeness. Nonetheless, it is sufficient for a fetch operation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the schema already documents both parameters. The description adds minimal extra meaning beyond the schema, just mentioning cross-platform support. Base score of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses the specific verb 'Fetch' and the resource 'system services', clearly stating the action and target. It also notes cross-platform support, distinguishing it from sibling tools like scan_system_logs or get_system_metrics.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage when service status is needed but provides no explicit guidance on when to use this tool versus alternatives, nor does it mention prerequisites or exclusion criteria.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Monitor system performance indicators: CPU, RAM, disk space, and hardware S.M.A.R.T health indicators. Supports Windows and Linux.
| Name | Required | Description | Default |
|---|---|---|---|
| targetHost | No | Optional remote target host (e.g., user@hostname) | |
| checkSmartHealth | No | If true, attempts to query drive health via smartctl |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the burden. It states it monitors performance indicators and OS support, but does not disclose potential side effects (assumed read-only), prerequisites (e.g., smartctl for S.M.A.R.T.), or error scenarios. Adequate but not rich.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences with no extraneous information. The first sentence immediately states the tool's purpose and scope. Highly efficient and front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple monitoring tool with 2 optional parameters and no output schema, the description covers the main points (metrics collected, OS support). Could optionally mention output format, but is largely complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% (2 parameters both described). The description adds context about what is monitored (CPU, RAM, disk, S.M.A.R.T.) which maps to checkSmartHealth, but does not elaborate on parameter usage beyond the schema. Baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly specifies the tool monitors system performance indicators (CPU, RAM, disk space, S.M.A.R.T. health) and supports Windows and Linux. It distinguishes itself from siblings like scan_system_logs or get_service_states which focus on logs or service states.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Usage is implied through the description (monitoring performance), but no explicit when-to-use or when-not-to-use guidance is given. No alternatives among siblings are mentioned.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Fetch and parse system/service logs locally. Summarizes errors and warning patterns to prevent token usage bloat.
| Name | Required | Description | Default |
|---|---|---|---|
| service | No | Filter by specific systemd service name (e.g. "nginx", "docker") | |
| maxLines | No | Maximum number of log lines to process locally before aggregation | |
| targetHost | No | Optional remote target host (e.g., user@hostname) | |
| filterKeywords | No | Case-insensitive keywords to filter logs |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description bears full responsibility for behavioral disclosure. It states logs are summarized locally and to prevent token bloat, but does not specify the output format, if it is read-only, or if any state is modified. The summarization mechanism (e.g., aggregation, suppression of details) is unclear.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences: one for action and one for purpose. It is concise and front-loaded, with no extraneous information. Slightly more detail on output could improve it.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With 4 parameters, no output schema, and no annotations, the description lacks completeness. It does not describe what the agent receives (summary format) or any side effects. More context is needed for the agent to use the tool correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents all parameters. The description adds little beyond the schema; for example, it mentions 'systemd service names' which aligns with the 'service' parameter. Baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Fetch and parse'), resource ('system/service logs'), and purpose ('summarizes errors and warning patterns to prevent token usage bloat'). It effectively distinguishes itself from siblings like 'get_service_states' or 'analyze_root_cause' by focusing on log summarization.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies use for log summarization to avoid token bloat, but lacks explicit guidance on when to use instead of siblings, prerequisites, or exclusions. No 'when-not' or alternative tools are mentioned.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v1.0.0analyze_root_causecheck_repair_strategy_limitget_platform_capabilitiesget_service_statesget_system_metricsscan_system_logsEach tool has a clearly distinct purpose: checking repair limits, querying platform capabilities, scanning logs, retrieving service states, monitoring metrics, and performing root cause analysis. No two tools overlap in functionality.
All tools follow a consistent verb_noun pattern in snake_case (e.g., check_repair_strategy_limit, get_platform_capabilities). No mixing of conventions or ambiguous names.
With 6 tools, the set is well-scoped for a system monitoring and analysis server. It covers core operations without excess or deficiency.
The tool set effectively covers analysis and observation (logs, metrics, services, platform capabilities, root cause). However, it lacks repair or remediation tools (e.g., apply fix, restart service), which is a minor gap given the 'repair' focus in the first tool's name.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Production-readiness for your AI coding agents.
Deterministic runtime safety for AI agents: scan PII, gate tool actions, verify LLM output.
Let AI operate servers without SSH. Choose actions, approve risky changes, and audit every step.
Agent-native security, trust, reliability, data and procurement tools for AI workflows.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/AnanthanarayanKV/SentinalMCP'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
