Vlm Code Context MCP Server
vlm-code-context-mcp
npm
v2.4.0
Published by velimirmueller — no publish provenance, so origin is unverified, but the source is public: the repository link below is self-declared yet readable, so you can inspect the code before adopting it.
AI-powered virtual IT department via MCP — codebase intelligence, 9-agent scrum team, sprint management, and React dashboard in one npm package
The grade answers one question — how safe is this server for you to adopt — so it is computed in two auditable stages. Nothing below is an opinion or an LLM's guess; every line is a real term the deterministic engine applied, and the same input always yields the same number.
1. Threat score — 100 − 1.2 = 99. What the published surface and source actually contain:
| Points | What was found | Category |
|---|---|---|
| −1.2 | Package runs install-time scripts MTC-SUP-010 | supply-chain |
2. Client adoption risk — 99 − 7 = 92. Three small, subtract-only factors that reflect your risk in adopting it — a clean scan proves less on a powerful, unverified or barely-inspectable package, so the grade says so plainly:
| Points | Adoption-risk factor |
|---|---|
| −6 | capability blast radius (high) — client exposure if the model is manipulated |
| −1 | publisher verification (public source) — no provenance, but the source is public and inspectable |
Capability observations and info notes are shown under Findings but never scored.
Open any row's finding below for the file, line and evidence behind a deduction.
Findings 8
high Read-and-egress in one tool: "export_to_file"MTC-FLOW-003
Tool "export_to_file" can both read sensitive data and send data to an external destination. Even without an explicit untrusted-input leg, this is a single-call data-exfiltration path if the model is ever manipulated.
Fix: Separate reading from sending; require explicit user confirmation before egress of file/secret contents.
Location: flow export_to_file
high Sensitive-source and external-sink co-existMTC-FLOW-004
Tools that read sensitive data ([dump_database, restore_database, export_to_file, import_from_file, get_token_usage]) and tools that can send data out ([export_to_file, request_user_input]) are exposed together. An agent can move private data to the sink.
Evidence: sources [dump_database, restore_database, export_to_file, import_from_file, get_token_usage] → sinks [export_to_file, request_user_input]
Fix: Keep secret-reading and egress capabilities on separate, separately-approved servers.
Location: flow dump_database → request_user_input
high Shell/command execution in server code (dist/dashboard/dashboard.js)MTC-SRC-002
In the server's implementation (`dist/dashboard/dashboard.js:464`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: { execSync } = require("child_process"); try { buildHash = execSync("git rev-parse --short HEAD", {
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/dashboard/dashboard.js
high Shell/command execution in server code (dist/scrum/defaults.js)MTC-SRC-002
In the server's implementation (`dist/scrum/defaults.js:162`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: { execSync } from "node:child_process"; /** * Pre-build check shared by the reset_* tools: ensure dist/ matches src/ be
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/scrum/defaults.js
high Shell/command execution in server code (dist/scrum/frontend-playbook.js)MTC-SRC-002
In the server's implementation (`dist/scrum/frontend-playbook.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: ecFileSync } from "node:child_process"; import { getEnabledSkillSets, SKILL_SETS } from "./skill-sets.js"; /** Read one
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/scrum/frontend-playbook.js
low Shell/command execution in packaging/dev tooling (scripts/make-legacy-fixture.mts)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`scripts/make-legacy-fixture.mts:13`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: ecFileSync } from "node:child_process"; import fs from "node:fs"; import os from "node:os"; import path from "node:path"
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server scripts/make-legacy-fixture.mts
low Shell/command execution in packaging/dev tooling (scripts/sync-skills.mjs)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`scripts/sync-skills.mjs:7`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: ecFileSync } from "node:child_process"; import { fileURLToPath, pathToFileURL } from "node:url"; const DEFAULT_REPO = "
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server scripts/sync-skills.mjs
low Package runs install-time scripts (postinstall)MTC-SUP-010
"vlm-code-context-mcp" executes postinstall script(s) at install time. An install hook runs at install time; most are routine build/setup, but review what it does before trusting it.
Evidence: node scripts/postinstall.js
Fix: Review the scripts; install with --ignore-scripts where possible and vet what they do.
Location: package vlm-code-context-mcp
Tools 98
Each tool and what it can reach — statically extracted from the published source.
export_to_filereads sensitive datanetwork egressdump_databasereads sensitive dataget_token_usagereads sensitive dataimport_from_filereads sensitive datarequest_user_inputnetwork egressrestore_databasereads sensitive dataacknowledge_ticket_changesno sensitive capabilityadd_dependencyno sensitive capabilityadd_retro_findingno sensitive capabilityadd_tagno sensitive capability
Show 88 more tools ↓
advance_sprintno sensitive capabilityanalyze_retro_patternsno sensitive capabilitycreate_blockerno sensitive capabilitycreate_discoveryno sensitive capabilitycreate_epicno sensitive capabilitycreate_milestoneno sensitive capabilitycreate_sprintno sensitive capabilitycreate_ticketno sensitive capabilityexecuteno sensitive capabilityexport_sprint_reportno sensitive capabilityfind_symbolno sensitive capabilitygenerate_vision_animationno sensitive capabilityget_agentno sensitive capabilityget_audit_trailno sensitive capabilityget_backlogno sensitive capabilityget_burndownno sensitive capabilityget_changesno sensitive capabilityget_dependency_graphno sensitive capabilityget_discovery_coverageno sensitive capabilityget_file_contextno sensitive capabilityget_mood_trendsno sensitive capabilityget_onboarding_statusno sensitive capabilityget_project_statusno sensitive capabilityget_resume_stateno sensitive capabilityget_skillno sensitive capabilityget_sprintno sensitive capabilityget_sprint_configno sensitive capabilityget_sprint_instructionsno sensitive capabilityget_sprint_playbookno sensitive capabilityget_sprint_summaryno sensitive capabilityget_ticketno sensitive capabilityget_time_reportno sensitive capabilityget_user_responseno sensitive capabilityget_velocity_trendsno sensitive capabilityhandle_run_kickoffno sensitive capabilityhandle_run_retrono sensitive capabilityhandle_run_reviewno sensitive capabilityhealthno sensitive capabilityindex_directoryno sensitive capabilitylink_discovery_to_ticketno sensitive capabilitylink_ticket_to_epicno sensitive capabilitylink_ticket_to_milestoneno sensitive capabilitylist_agentsno sensitive capabilitylist_decisionsno sensitive capabilitylist_discoveriesno sensitive capabilitylist_epicsno sensitive capabilitylist_recent_eventsno sensitive capabilitylist_retro_findingsno sensitive capabilitylist_sprintsno sensitive capabilitylist_tagsno sensitive capabilitylist_ticketsno sensitive capabilityload_phase_contextno sensitive capabilitylog_bugno sensitive capabilitylog_decisionno sensitive capabilitylog_eventno sensitive capabilitylog_timeno sensitive capabilitylog_token_usageno sensitive capabilityplan_sprintno sensitive capabilityqueryno sensitive capabilityrecord_moodno sensitive capabilityremove_dependencyno sensitive capabilityremove_tagno sensitive capabilityreset_agentsno sensitive capabilityreset_skillsno sensitive capabilityreset_sprint_processno sensitive capabilityresolve_blockerno sensitive capabilityrun_onboardingno sensitive capabilitysearch_filesno sensitive capabilitysearch_scrumno sensitive capabilitysend_claude_outputno sensitive capabilitysend_claude_stepno sensitive capabilitysend_step_progressno sensitive capabilityset_change_reasonno sensitive capabilityset_descriptionno sensitive capabilityset_directory_descriptionno sensitive capabilitysnapshot_sprint_metricsno sensitive capabilitystart_sprintno sensitive capabilitysync_scrum_datano sensitive capabilitytriage_retro_findingno sensitive capabilityupdate_agentno sensitive capabilityupdate_discoveryno sensitive capabilityupdate_epicno sensitive capabilityupdate_milestoneno sensitive capabilityupdate_skill_setsno sensitive capabilityupdate_sprintno sensitive capabilityupdate_sprint_configno sensitive capabilityupdate_ticketno sensitive capabilityupdate_visionno sensitive capability
Toxic flows 2
Cross-tool combinations that form a data-exfiltration primitive (untrusted input → sensitive source → external sink).
What this scan could not see
- Tools were statically extracted from the published source (98 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Versions 2
Scan history per published version. The engine is deterministic — the same version always yields the same score, so a changed score means the package itself changed.
| Version | Score | Findings | Engine | Scanned |
|---|---|---|---|---|
v2.4.0 latest |
A 92/100 | 8 | 1.13.0 | 2026-09-07 |
v2.3.0 |
A 92/100 | 4 | 1.13.0 | 2026-09-01 |
Embed this score
Show this server's live Trust Score in your README, docs or website. The badge is served straight from the registry and updates automatically after every rescan — no API key needed. It links back to this page, so anyone who sees the grade can also read the findings behind it instead of taking a number on faith.
Verify this score yourself
The score above is reproducible: the same package version always yields the same result. Run it locally or over the free API — no account, no LLM, fully deterministic.
npx mcptrustchecker scan vlm-code-context-mcp --onlineMore in Developer Tools
Hotel booking MCP server — 300K+ properties, real confirmation numbers, loyalty programs. Builders monetize every booking via Stripe Connect. The first MCP server that completes real hotel reservations inside AI conversations.
Manage AdGuard Home through AI assistants
Read-only Azure DevOps for MCP clients using only your existing browser session — no PAT, no Azure CLI. Browse work items, pull requests, comments, attachments and Artifacts feeds across every project, repo and feed you can access.
MCP server for Adobe Experience Manager Assets integration development
Servidor MCP para el tiempo oficial de España (API pública OpenData de AEMET). Predicción, observación y avisos como herramientas MCP tipadas.
A standalone MCP stdio bridge for Affinity by Canva's local MCP SSE server.