io.github.ohad6k/viberaven
NPM · @VIBERAVEN/MCP
Checks a Vercel + Supabase repo before deploy: RLS, grants, leaked keys, webhooks. Runs locally.
Available components
Last scanned (19 hours ago)
Scores higher than 55% of npm servers
Category Security
How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →
Supply Chain Security98
- No malware found by supply-chain analysis.Pass
- No known CVEs affecting this package version or its production dependencies.Pass
- No install/post-install scripts declared.Pass
- 30 of 92 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency19
- Repository check failed: no source repository is declared. See how to fix → View diagnostics → Fail
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- Clear OSI-approved license (MIT).Pass
- Actively maintained (last published 1 days ago).Pass
- Disclosure check failed: no security disclosure policy was found in the source repository. See how to fix → Fail
Schema Quality & AI Usability69
- AI-judged instruction clarity (good).Pass
- Context-footprint check failed: tool/resource definitions use about 3359 tokens (~129/item across 26 items; 26 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management100
- No destabilizing schema changes in the last 30 days.Pass
Tool Coverage90
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 67% of tool parameters carry a description.Partial
- Structured output schemas are declared (4% of tools); any adoption earns full credit.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 26 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 27 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 31 Oct 26 +15
- Malware scan: unverified → pass ▲ security
- 30 Oct 26 −22
- Malware scan: pass → unverified ▼ security
- Source repository: pass → fail ▼ security
- Schema quality: 115 → 129 ▼ functional
- Tool coverage: 86% → 67% ▼ functional
- Security disclosure: unverified → fail ▼ functional
- First check of Tool coverage: 4 functional
- Schema quality: excellent → good functional
- Package version: 1.6.12 → 1.7.1 functional
- 29 Oct 26 0
- Security disclosure: fail → unverified ▼ functional
- 28 Oct 26 +15
- Malware scan: unverified → pass ▲ security
- 27 Oct 26 −9
- Malware scan: pass → unverified ▼ security
- Source repository: fail → pass ▲ security
- Stability: 0.97 → pass security
- Package version: 1.6.7 → 1.6.12 functional
- 25 Oct 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.
- 23 Oct 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.
- 21 Oct 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 77 to 80. That category is still filling its 30-day observation window: 23 days of observed history at the previous scan, 24 at this one. The score rises as the window fills, whether or not the server changes.
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 4 Nov 2026 · Analysed npm/@viberaven/mcp@1.7.1
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | npm |
Background: How many MCP packages publish verified provenance →
Dependencies 92 packages
| Packages resolved | 92 |
|---|---|
| Stale | 30 |
| Tree resolution | Complete |
Background: SBOMs and build attestations, explained →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
provider_playbook Get a provider playbook ~99
Plan setup and integration before implementing a feature that touches a provider. Returns versioned setup guidance, gotchas, dated changes, integration seams, existing repo checks and planned live checks with official sources. Available packs: supabase. Limits: guidance only; does not connect to providers, read your project, run checks or verify that a feature works.
| Name | Type | Req | Description |
|---|---|---|---|
| provider | string | yes | Provider id, for example supabase. Unsupported ids return the available packs. |
| Name | Type | Req | Description |
|---|---|---|---|
| nextStep | object | yes | – |
| pack | object | yes | – |
| packHash | string | yes | – |
| packVersion | string | yes | – |
| provider | string | yes | – |
| verification | object | yes | – |
No examples provided.
viberaven_actions ~60
Read the current chat-native action surface from .viberaven/actions.json. Use this to list stable, referenceable action IDs the user can verify or refer to by name.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Project root. Defaults to current working directory. |
No output schema declared.
No examples provided.
viberaven_audit ~105
Run local Vercel + Supabase production checks: table RLS presence, policy quality, view/function boundaries, service-role boundaries, and Vercel pooler usage. Call this when the project uses Supabase or Vercel and you need a focused infra audit beyond the general readiness check. Pass json: true for machine-readable output.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Project root. Defaults to current working directory. |
| json | boolean | – | Return JSON output. |
No output schema declared.
No examples provided.
viberaven_check_readiness ~148
Useful before an AI-built app on Vercel and Supabase launches or is handed off. Runs the local `viberaven check` (no login, no API key, no telemetry), which reads the repository for launch gaps (missing RLS in migrations, leaked secrets, unbounded queries, unsigned webhooks) and writes a prioritized task list to .viberaven/agent-tasklist.md plus a verdict to .viberaven/gate-result.json. This is advice, not a gate: the user decides when to ship. It reads repo files and does not query the live database.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Project root. Defaults to the MCP server cwd. |
No output schema declared.
No examples provided.
viberaven_clean_plan ~69
Write a non-destructive cleanup plan for generated .viberaven artifacts and logs. Call this when the .viberaven directory has grown stale. Does not delete anything; it only produces a reviewable plan.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Project root. Defaults to current working directory. |
No output schema declared.
No examples provided.
viberaven_context_map ~63
Refresh and return .viberaven/context-map.json, a compact summary of the last scan for the next agent step. Use this to load minimal context before acting on a gap.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Project root. Defaults to the MCP server cwd. |
No output schema declared.
No examples provided.
viberaven_gate_result ~85
Run a fresh local scan (`viberaven --json`) and return the verdict it writes to .viberaven/gate-result.json (gate.status is clear, warning or not_clear). It rewrites the .viberaven artifacts; it does not read an older result without scanning.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Project root. Defaults to the MCP server cwd. |
No output schema declared.
No examples provided.
viberaven_heal_apply ~159
Apply a guarded repo-code heal recipe for a supported gap. Only call this after viberaven_heal_plan. Pass yes: true to skip confirmation. The rls_disabled recipe writes a migration that enables row level security with no policies, so browser reads of those tables return no rows until policies exist: call it only after the user agrees. Its task leaves out yes: true; pass yes: true only once the user agrees. Returns the applied change summary, with any warnings.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | – |
| gap | string | – | Gap ID from .viberaven/agent-tasklist.md. |
| target | string | – | Target file path for the heal. |
| yes | boolean | – | Skip confirmation for apply. |
No output schema declared.
No examples provided.
viberaven_heal_plan ~144
Write a non-destructive heal plan (.viberaven/heal-plan.md) for a specific gap or target file, to review before viberaven_heal_apply. The plan names the target and the verify command. For rls_disabled it also shows the new migration it would write and its SQL, or why it will not write one; other gaps get no patch preview. Does not modify source files.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | – |
| gap | string | – | Gap ID from .viberaven/agent-tasklist.md. |
| target | string | – | Target file path for the heal. |
| yes | boolean | – | Skip confirmation for apply. |
No output schema declared.
No examples provided.
viberaven_heal_prompt ~93
Return an agent-ready prompt that fixes a specific gap. Use this when you want a coding agent to apply the fix itself rather than using the guarded recipe.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | – |
| gap | string | – | Gap ID from .viberaven/agent-tasklist.md. |
| target | string | – | Target file path for the heal. |
| yes | boolean | – | Skip confirmation for apply. |
No output schema declared.
No examples provided.
viberaven_init_rules ~106
Install bounded VibeRaven agent-instruction rules into native AI instruction files (CLAUDE.md / AGENTS.md / .cursor/rules), so coding agents know when a VibeRaven pass applies. Use dryRun: true to preview first.
| Name | Type | Req | Description |
|---|---|---|---|
| agents | string | – | Comma-separated agent targets, or all. |
| cwd | string | – | Project root. Defaults to current working directory. |
| dryRun | boolean | – | Preview changes without writing files. |
No output schema declared.
No examples provided.
viberaven_live_disconnect ~93
Local candidate feature: disconnect the saved live step and remove its locally retained provider credential and evidence. Does not alter provider settings or unrelated MCP credentials. Unfinished account reservations expire. Local checks stay free.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Local project directory. Defaults to the MCP working directory. |
| stepId | string | yes | The stepId returned by live_start; preserve it through sign-in, payment and retries. |
No output schema declared.
No examples provided.
viberaven_live_resume ~152
Local candidate feature: resume the SAME saved live step through account connection, allowance and metadata verification. Set approvedRead=true only after the person explicitly approves the named Supabase project/schema catalog read. Show any returned action URL and wait for the person; never approve sign-in or payment for them. Credentials stay local and are never tool arguments. Vercel stays blocked; local checks stay free.
| Name | Type | Req | Description |
|---|---|---|---|
| approvedRead | boolean | – | True only after explicit human approval of this saved provider/project/schema read. |
| cwd | string | – | Local project directory. Defaults to the MCP working directory. |
| stepId | string | yes | The stepId returned by live_start; preserve it through sign-in, payment and retries. |
No output schema declared.
No examples provided.
viberaven_live_start ~201
Local candidate feature: save a scoped metadata-verification request without provider reads or payment. Supabase catalog reads require a VibeRaven account and explicit human consent on resume. Vercel is deliberately blocked before network access. Local repository checks stay free without an account. Never pass credentials as tool arguments.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Local project directory. Defaults to the MCP working directory. |
| projectRef | string | yes | Exact provider project ID: 20 lowercase letters for Supabase, or prj_ followed by 8–80 alphanumeric characters for Vercel. |
| provider | string | yes | – |
| schemas | array | – | 1–8 lowercase application schemas; defaults to public. Excludes pg_, auth, storage, vault and information_schema. No application-row reads. |
| teamId | string | – | Optional Vercel team_ ID with 8–80 alphanumeric characters. Vercel transport is currently blocked. |
No output schema declared.
No examples provided.
viberaven_live_status ~91
Local candidate feature: read the saved live-step receipt without provider reads, payment or another live-step charge. Evidence is historical and scoped, not proof of application security. Local checks stay free.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Local project directory. Defaults to the MCP working directory. |
| stepId | string | yes | The stepId returned by live_start; preserve it through sign-in, payment and retries. |
No output schema declared.
No examples provided.
viberaven_rls_test ~176
After reviewing a Supabase migration or ownership policy, optionally execute already reviewed local permissions scenarios for owner, other_user and anon. Returns Pass, Fail or Unknown per cell with the original scoped evidence. This is an experimental PGlite imitation, not live Supabase proof. The optional PGlite dependency must already be installed next to the matching CLI used by this local MCP host; missing setup returns the CLI setup guidance as a tool error. No SQL, credentials, provider calls or dependency installation are accepted.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Project root. Defaults to the MCP server working directory. |
| matrix | string | yes | Path to an already reviewed permissions JSON file, relative to cwd or absolute. Expectations must describe intended access. |
| migrations | string | – | Local migrations directory. Defaults to supabase/migrations under cwd. |
No output schema declared.
No examples provided.
viberaven_strict_gate ~85
Run `viberaven --strict --json` and return the verdict from .viberaven/gate-result.json (gate.status is clear, warning or not_clear). Useful before a deploy or CI run when the user wants the verdict as JSON. The user decides whether to ship.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Project root. Defaults to the MCP server cwd. |
No output schema declared.
No examples provided.
viberaven_validate_npm_package ~182
Look up npm package names on the public registry (registry.npmjs.org). Returns not_found when the registry has no such package or cannot be reached, suspicious when the registry answers with an error other than 404, and suspicious when the package was first published less than 14 days ago and its name is within 2 character edits of a popular package (a typosquat signal), or when it has an empty description, no listed maintainers and 3 or fewer published versions. Anything else is ok, which only means none of these signals fired; it is not a security review of the package. Useful when an agent is about to add a dependency the project has not used before.
| Name | Type | Req | Description |
|---|---|---|---|
| packageName | string | – | Single npm package name to validate. |
| packageNames | array | – | Multiple package names to validate in one call. |
No output schema declared.
No examples provided.
viberaven_verify ~72
Re-run the local VibeRaven check after you apply fixes. Call this once per batch of fixes (not after every edit) to update .viberaven/gate-result.json and confirm whether the gate moved toward clear.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | Project root. Defaults to the MCP server cwd. |
No output schema declared.
No examples provided.
viberaven_verify_action ~69
Verify or explain the lifecycle state of a stable VibeRaven action ID (format VR-A<number>). Use this when the user references a specific action and you need its current state.
| Name | Type | Req | Description |
|---|---|---|---|
| actionId | string | yes | Action ID like VR-A1. |
| cwd | string | – | – |
No output schema declared.
No examples provided.
viberaven_work_prepare ~213
Before implementing a local Node feature, freeze reviewed expected behavior and actual dirty/untracked baseline. Supports module comparisons, sequential HTTP requests to an isolated local server factory, and frozen Node preservation tests. No login or network; no commands or imported proof. MCP plans are limited to 4800 UTF-8 bytes; larger plans use the CLI JSON file (65536 bytes). No browser, database or live provider proof. Changed plans require a new task.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | – |
| taskId | string | yes | – |
| verification | object | yes | Reviewed local observations, not commands or imported proof. Runtime also bounds JSON depth/UTF-8 bytes, including HTTP request/expected bodies to 8192 bytes, rejects unknown condition IDs, and restr… |
No output schema declared.
No examples provided.
viberaven_work_report ~98
Export a bounded local draft proof report with Markdown from current integrity-checked verification receipts. Read-only: no login, network, Docker, agent or GitHub execution. Failed and unknown evidence remain explicit; agent success is not proof. Not authoritative CI or a completed live PR pilot. Review redacted text before sharing; redaction is not a complete secret scanner.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | – |
| taskId | string | yes | – |
No output schema declared.
No examples provided.
viberaven_work_resume ~160
Resume the SAME saved task through account sign-in, quota or payment. Set approvedContract=true only after explicit human approval of its exact acceptance, must-not-break conditions and declared scope. A draft can be upgraded with contract before any job exists; its request must stay identical. Never authorize sign-in or payment for the person. Execution state is separate from independent verification observations; agent success is not condition proof. Scope declarations are not an enforced sandbox.
| Name | Type | Req | Description |
|---|---|---|---|
| approvedContract | boolean | – | – |
| contract | object | – | Declarative task definition, not executable commands or evidence. Runtime also enforces a 4800-byte UTF-8 limit and condition IDs unique across both lists. |
| cwd | string | – | – |
| taskId | string | yes | – |
No output schema declared.
No examples provided.
viberaven_work_start ~153
Save a versioned Task contract and run a free local check without login, network, edits or payment. Prefer contract; request alone saves a draft with unknown acceptance. No credentials or shell command fields. Review the returned contract before approving resume. Resume shares its short request and redacted results with the account; the full definition stays local. Provider and non-local contracts are planning only.
| Name | Type | Req | Description |
|---|---|---|---|
| access | string | yes | – |
| cli | string | yes | – |
| contract | object | – | Declarative task definition, not executable commands or evidence. Runtime also enforces a 4800-byte UTF-8 limit and condition IDs unique across both lists. |
| cwd | string | – | – |
| request | string | – | – |
No output schema declared.
No examples provided.
viberaven_work_status ~79
Read the saved Task contract, execution state and independent verification receipts locally without account, network, edits or payment. Current worktree changes invalidate checked proof; missing observations stay not_checked. Checked conditions separately report passed/failed outcomes. Historical agent results do not verify conditions.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | – |
| taskId | string | yes | – |
No output schema declared.
No examples provided.
viberaven_work_verify ~101
Run or resume the SAME frozen plan using the locally available pinned Docker image only. Set approvedVerification=true only after human review/approval. Network is disabled and no dependencies are installed. No login or billing. Missing conditions, dependencies or Docker remain not_checked; completed receipts are reused for the identical candidate. Coding-agent success is not proof.
| Name | Type | Req | Description |
|---|---|---|---|
| approvedVerification | boolean | yes | – |
| cwd | string | – | – |
| taskId | string | yes | – |
No output schema declared.
No examples provided.
How do I install the io.github.ohad6k/viberaven MCP server?
io.github.ohad6k/viberaven runs locally as an npm package, launched with npx -y @viberaven/mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
npm · @viberaven/mcp
claude mcp add ohad6k-viberaven -- npx -y @viberaven/mcp
What is the io.github.ohad6k/viberaven MCP server?
io.github.ohad6k/viberaven is an MCP server listed in the public MCP registry as io.github.ohad6k/viberaven. Checks a Vercel + Supabase repo before deploy: RLS, grants, leaked keys, webhooks. Runs locally. This page covers its npm package (@viberaven/mcp).
Is the io.github.ohad6k/viberaven MCP server safe to use?
io.github.ohad6k/viberaven scores 77 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 4 November 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.
What tools does the io.github.ohad6k/viberaven MCP server expose?
io.github.ohad6k/viberaven exposes 26 tools: viberaven_check_readiness, viberaven_rls_test, viberaven_verify, viberaven_audit, viberaven_init_rules, and 21 more. Their descriptions and schemas cost roughly 3,056 tokens of context every time the server is loaded.
Is the io.github.ohad6k/viberaven MCP server still maintained?
io.github.ohad6k/viberaven is still listed as active in the MCP registry. We last reached this channel on 4 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.
What licence is the io.github.ohad6k/viberaven MCP server under?
io.github.ohad6k/viberaven declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.