remote-access-mcp (AmirAliManzar/remote-access-mcp) is an MCP server listed on the M8ven Trust Index. It scores 49 out of 100, grade D. It declares 122 tools. No publisher has claimed this listing.
Turn any Linux server into an AI-agent-accessible machine via MCP. ChatGPT, Claude, and Grok connect over HTTPS and control files, shell, git, and more — securely, with zero Python dependencies.
Caution. Specific findings reduced this grade. They are listed on the page. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.
How we verified
Verified is a snapshot. Live keeps it current, and builds your track record.
⚡ Connect GitHub → continuous verification on every pushwhy connect →Who stands behind it
AmirAliManzar
Source: github_repo_search
Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find. Or connect your repo for our deepest verification, Live Monitored: read-only, revoke anytime. What we access →
Install from
The grade above is for the source repository. Registries can serve a different version, so we mark the ones we were not able to read.
These names and descriptions are the publisher's own, read from the source code. We print them as written. Our assessment is the findings above, not this list.
agent_profilesList specialized task-agent profiles and their allowed capability scopes.
automation_createCreate a persistent token-isolated automation rule. Trigger can be interval, tool, webhook, file, or health; actions execute through the normal policy gate.
automation_listList automations belonging only to the current token.
automation_statusShow status and run counters for one owned automation.
automation_enableEnable or disable one of the current token automations.
automation_triggerManually trigger an owned automation rule after evaluating its conditions.
automation_deleteDelete one of the current token automations.
security_analysisRun a bounded read-only security posture analysis for the current token.
autonomy_checkEvaluate whether an action risk may execute at a requested autonomy level.
recovery_rule_createCreate a bounded token-isolated self-healing rule for one known failure type.
recovery_rule_listList recovery rules owned by the current token.
recovery_rule_deleteDelete an owned recovery rule.
recovery_incidentsList token-isolated self-healing incidents.
recovery_triggerRun one owned recovery rule once, subject to cooldown, attempt limits, autonomy, and normal tool policy.
recovery_resetReset the bounded attempt counter of an owned recovery rule.
browser_openOpen a public web page in a headless browser and return its title, URL, and visible text. SSRF-guarded.
browser_extractExtract structured links and metadata from a public page using a headless browser.
browser_screenshotCapture a screenshot of a public web page to a policy-allowed local path.
change_set_beginBegin a file change set. Mutating filesystem tools can attach to it so rollback also handles created and deleted paths.
change_set_addRegister a path in a change set by capturing its current state. Missing paths are tracked so rollback removes files created later.
change_set_statusShow a change set and its captured pre-mutation states.
change_set_commitCommit a change set. The recorded backups remain until explicit cleanup.
change_set_rollbackIdempotently restore every captured path. Created paths are removed; deleted or modified paths are restored. A crash during rollback leaves rolling_back state so the operation can be resumed.
context_statsShow context-engine mode, cache, memory and byte-saving statistics for the current token.
context_memoryReturn recent compact task memory for the current token only.
context_snapshotSave a context snapshot for the current token and return a short snapshot id.
context_diffCompare text with a saved context snapshot and return a compact line diff.
context_budgetCompact arbitrary text to an explicit context-character budget.
context_clearClear this token's local context cache and task memory.
database_queryRun a database query using an installed mysql, psql, or redis-cli client. Credentials are read from a named environment variable and never returned.
database_schemaInspect database schema using an installed mysql or psql client. Credentials stay in an environment variable.
system_diagnosticsCollect a structured system diagnostic snapshot: CPU, memory, load, uptime and disk summary.
diagnose_serviceDiagnose a service using status and recent journal/log output.
list_directoryList a directory. Policy-checked.
read_fileRead a UTF-8 text file. Policy-checked.
write_fileCreate or overwrite a file with UTF-8 text content. Policy-checked.
edit_fileReplace exact text inside a file. Policy-checked.
delete_pathDelete a file or a directory tree. Policy-checked locally. Destructive.
search_codeRecursive regex search under a directory. Policy-checked. Returns file:line:match.
file_infoStat a file or directory: size, timestamps, permissions, owner. Policy-checked locally.
gitRun git commands in a repository (verbs whitelisted, options validated). Policy-checked.
http_requestOutbound HTTP request. SSRF-guarded: private/metadata ranges refused. 20s timeout.
port_checkCheck if a TCP port is reachable on an external host. Refuses internal ranges. 5s timeout.
infra_probeProbe availability and versions of Docker, kubectl, and cloudflared without changing the host.
docker_psList Docker containers. Read-only and shell-free.
docker_inspectInspect one Docker container by name or ID.
docker_logsRead recent Docker container logs.
docker_actionStart, stop, or restart one Docker container. Requires mutation permission.
kubernetes_getRead Kubernetes resources with kubectl get. No arbitrary kubectl flags.
kubernetes_describeDescribe a Kubernetes resource using a fixed kubectl command.
kubernetes_logsRead recent logs from a Kubernetes pod/container.
cloudflare_statusInspect cloudflared version and configured tunnel visibility. Never returns credentials.
context7project_profileGet the persisted developer profile for a project.
project_profile_listList project profiles belonging to the current token.
project_profile_setPersist a project profile under RAMCP data, isolated to the current token.
impact_analysisBuild a lightweight reverse dependency graph and identify source files impacted by changes.
git_intelligenceRead-only Git intelligence: status, recent history, changed files, branches and remotes.
github_repoRead a GitHub repository summary using owner/name.
github_issuesList GitHub issues for a repository. Read-only.
github_pull_requestRead a GitHub pull request and its changed files.
sentry_projectsList Sentry projects available to the configured token.
sentry_issuesList recent Sentry issues for an organization/project.
sentry_issueInspect one Sentry issue.
developer_context_statusReport Codebase Memory isolation, Context7 proxy and Context Mode local compatibility.
run_backgroundRun a shell command asynchronously in the local worker pool. Returns a job id.
job_statusGet status and recent output of a background job.
job_outputRead accumulated stdout/stderr from a background job.
job_listList background jobs belonging to the current token.
job_cancelCancel a queued or running background job.
run_parallelQueue multiple shell commands for parallel execution by the shared worker pool. Concurrency is bounded globally.
tail_logsLast N lines of a text log file. Policy-checked. No follow (MCP is request/response).
search_logsRegex search in a log file, newest-first, with context lines. Policy-checked.
journalQuery system logs for a service. journalctl on Linux, log show on macOS, Get-EventLog on Windows.
health_watchStart a persistent periodic health watcher for CPU/memory thresholds. Alerts are emitted through configured webhooks and survive gateway restart.
health_statusList active persistent health watchers for the current token.
health_stopStop a persistent health watcher.
environment_inspectShow gateway environment variables with secrets masked. Read-only.
nginx_inspectInspect the local nginx configuration: parsed vhosts, enabled sites, upstreams. Read-only.
package_listList installed packages. Auto-detects apt / brew / winget / choco, falls back to global npm.
package_installInstall a package using the platform manager (apt/brew/winget/choco) or npm -g for scoped names.
package_removeRemove a package. Refuses packages the gateway itself depends on.
create_task_planCreate a multi-step task plan the AI can track. Steps can be marked done via task_status.
task_statusShow a task plan, or mark steps done by index (1-based).
workspace_snapshotSnapshot files (copy) before a risky change. Roll back with rollback_changes. Policy-checked.
rollback_changesRestore files from a snapshot. Policy-checked.
plugin_listList installed Remote Access MCP plugins and integrity verification status.
plugin_installInstall a local plugin after validating its manifest, filesystem tree and integrity fingerprint.
plugin_removeRemove an installed local plugin and its integrity record.
list_allowed_pathsShow this token's path policy and shell flag.
allow_pathGrant this token access to a directory path.
deny_pathRevoke this token's access to a directory path.
shell_enabledCheck whether shell command execution is allowed for this token.
analyze_projectSummarize a project: languages, file counts, LOC, entry points, manifests, frameworks. Policy-checked.
project_health_checkQuick health check: git dirty state, huge files, missing README, TODO/FIXME density. Policy-checked.
system_resourceReturn the current system resource as structured JSON.
capability_discoverDiscover only the capabilities available to the current token. Use query/category/cost to keep context small.
capability_batchRun multiple independent read-only tool calls in parallel. Maximum 8 calls; mutating tools are rejected.
schedule_commandSchedule a shell command: one-shot at an ISO time, or recurring every N seconds. Shell must be enabled.
list_scheduled_tasksList all scheduled tasks with their next run times.
22 further tools are not listed here. The complete surface is in the source.
Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.
When Vitest UI server is listening, arbitrary file can be read and executed
Vitest: Path Traversal / Arbitrary File Read via @vitest/mocker Redirect Mock
ComSpecRAMCP_AUTONOMOUSRAMCP_AUTONOMOUS_CRITICALRAMCP_AUTONOMOUS_HIGH_RISKRAMCP_CODEBASE_LOG_LEVELRAMCP_CODEBASE_MEM_BUDGET_MBRAMCP_CODEBASE_ROOTRAMCP_CODEBASE_WORKERSRAMCP_CODEBASE_WORKSPACERAMCP_CONTEXT_MODERAMCP_TOOL_EXPOSURERAMCP_WORKERSXDG_CONFIG_HOMETool annotations
No tools have read-only/destructive annotations
Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.
All four hints declared on every tool
122/122 tools missing one or more hints — agent_profiles (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); automation_create (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); automation_list (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +119 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.
For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.
Destructive tools are labelled
6 tools perform destructive updates without destructiveHint — change_set_rollback deletes at line 65 (fs.rm(target, { recursive: true, force: true })); delete_path deletes at line 111 (fsp.rm(target, { recursive: true })); health_watch deletes at line 51 (fs.rmSync(lockPath(), { recursive: true, force: true }))
Add destructiveHint:true to any tool whose handler calls .delete(), .upsert(), .update(), unlink, rm, DELETE, DROP, REPLACE INTO, or any operation that overwrites existing data.
Descriptions match behaviour
10 tools describe read intent but their handlers mutate — database_schema (line 11: exec(bin,args,{env:{...process.env,MYSQL_PWD:engine==='mysql'?secret:process.env.MYSQL_PWD,PGPASSWOR…); git_intelligence (line 18: exec('git', args, { cwd: root, maxBuffer: 4 * 1024 * 1024 })); journal (line 90: exec(file, args, { env: childEnv(), windowsHide: true, maxBuffer: 8 * 1024 * 1024 }))
Rename the tool, rewrite the description, or move the side-effect into a separate clearly-named tool.
Tool test coverage
48/122 tools referenced in tests (39%)
Write tests that reference each tool by name so every tool has at least one test.
Shell command execution
1 call in production code run through a shell (src/server/run.ts:64)
Prefer library functions over shell-outs. If you must shell out, ensure all inputs are properly escaped.
Dev dependencies
1 critical/high in dev-only deps (does not ship to users)
Upgrade dev dependencies when convenient.
Tool description accuracy
9 tools have description/behavior mismatches: git_intelligence: description implies read-only but handler writes/deletes/executes; journal: description implies read-only but handler writes/deletes/executes; nginx_inspect: description implies read-only but handler writes/deletes/executes
Update tool descriptions to accurately reflect all capabilities — especially write, delete, or execute operations.
Claim the listing to review these findings one by one and send us a correction where you disagree, straight to the team. Claiming also means we tell you when the grade moves, and reach you first if we find anything urgent.
[](https://m8ven.ai/mcp/amiralimanzar/remote-access-mcp)?variant=verified from the URL.Vetting this one by hand? Tool Check is an MCP that scores other MCPs. Add it once and ask Claude, ChatGPT, or any MCP client to grade a server, surface CVEs, check the publisher, and suggest safer alternatives — before you install.
https://m8ven.ai/api/mcp/tool-check