Server providing Argo CD integration for MCP clients.
NEPHIS is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
What it does
Sub-6ms microVM sandboxes, zero-copy Postgres branching, and fast code execution in ap-south-1.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user nephis-mcp '--' npx -y nephis-mcp-server
# 2 — ask your agent for something these tools do
# thanaton_exec_command, thanaton_run_code, thanaton_spawn_sandbox and 72 more
NEPHIS costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts NEPHIS as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
This is a public server: you run it yourself and this marketplace is not in the path. You need Node.js — the current LTS (24) is the safe choice; npx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user nephis-mcp '--' npx -y nephis-mcp-serverReconnect, or start a new session, and the tools appear in the model’s tool list.
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
75 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
Execute an arbitrary shell command (bash, sh) inside an isolated, hardware-virtualized MicroVM sandbox. Returns stdout, stderr, exit code, and execution latency.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| command | string | yes | The shell command to execute inside the MicroVM (e.g., 'python3 -c "print(2+2)"', 'npm test', 'git status') |
| sandbox_id | string | no | Optional existing sandbox ID. If omitted, an ephemeral sub-6ms sandbox will be provisioned automatically. |
| timeout_ms | number | no | Maximum execution timeout in milliseconds (default: 15000) |
Directly run a standalone, single-file code snippet in an ephemeral, stateless microVM sandbox (automatically destroyed after execution). For stateful multi-step terminal interactions, filesystem persistence, or long-running processes, use 'thanaton_spawn_sandbox' followed by…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| language | string | yes | Programming language runtime |
| code | string | yes | The complete source code snippet to compile and execute |
| timeout_ms | number | no | Timeout in milliseconds (default: 10000) |
Provision a new hardware-isolated KVM MicroVM or Silicon isolate. Supports scale-to-zero, memory snapshots, and optional embedded PostgreSQL database.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| runtime | string | no | Runtime environment type (default: 'microvm') |
| memory_mb | number | no | Allocated RAM in MiB (default: 256) |
| with_db | boolean | no | Whether to bundle an isolated PostgreSQL database inside the sandbox |
List all active, sleeping, and managed MicroVM sandboxes across the current tenant.
No parameter schema on file.
Instantly wake a sleeping/hibernated MicroVM from NVMe memory snapshot in sub-6ms (<2ms with UFFD lazy-paging).
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sandbox_id | string | yes | The ID of the sleeping sandbox to wake |
Scale a MicroVM to zero (hibernates RAM to NVMe snapshot, 0% CPU consumption, $0 cost).
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sandbox_id | string | yes | The ID of the running sandbox to put to sleep |
Terminate processes, tear down TAP network interface, cgroups, and reclaim all sandbox resources.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sandbox_id | string | yes | The ID of the sandbox to destroy |
List all Thanaton serverless application projects and their live domains.
No parameter schema on file.
Create a new project/application deployment target in Thanaton.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | Project name |
| slug | string | no | Optional custom project slug |
| runtime | string | no | Application runtime (default: nodejs) |
| app_port | number | no | Internal HTTP listening port (default: 3000) |
Deploy code directly to a Thanaton project or application slot. Automatically boots the workload and routes public domain traffic to it.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_slug | string | yes | Target project slug or ID (e.g. mohit-aiops-cloud-intelligence) |
| code | string | no | Source code of the application entrypoint (e.g. FastAPI, Flask, Express, or Next.js code) |
| filename | string | no | Entrypoint filename (default: app.py for python, server.js for nodejs) |
| runtime | string | no | Runtime environment |
Retrieve real-time runtime stdout/stderr logs from a deployed Thanaton project workload.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_slug | string | yes | Target project slug or ID |
List, set, or delete project environment variables.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Target project ID |
| action | string | yes | Action to perform |
| key | string | no | Environment variable key |
| value | string | no | Variable value (for set action) |
| environment | string | no | Target environment |
Instant atomic rollback of production traffic to a previous deployment snapshot in 8.2ms.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Target project ID |
| deployment_id | string | yes | Target snapshot deployment ID to restore |
Promote staging preview deployment snapshot to 100% production traffic in 6.4ms.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Target project ID |
Inspect or configure edge Web Application Firewall (WAF) Attack Mode, rate limits, allowlist, and blocklist.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Target project ID |
| action | string | yes | Action to perform |
| attack_mode | boolean | no | Enable Under Attack Mode (cryptographic PoW challenge) |
| rate_limit_rps | number | no | Per-IP rate limit RPS threshold |
| allowlist | array | no | CIDR allowlist IP blocks |
| blocklist | array | no | CIDR blocklist IP blocks |
Inspect or configure Thanaton compute scaling: toggle between Serverless (scale-to-zero) and Always-On stateful execution, RAM allocation, and idle timeout.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Target project ID |
| action | string | yes | Action to perform ('get' or 'set') |
| always_on | boolean | no | Enable 24/7 Always-On stateful background execution (Indie/Pro) |
| idle_timeout_seconds | number | no | Inactivity seconds before scaling to zero (min 15s) |
| memory_mb | number | no | RAM allocation in MB (e.g. 256, 1024, 4096) |
| vcpu_count | number | no | vCPU count (e.g. 1, 2, 4) |
List active PostgreSQL database tenants, clusters, and copy-on-write branches with replication and fork telemetry.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | no | Optional tenant filter |
Fork an instant copy-on-write PostgreSQL database branch in 31ms at an exact WAL LSN without storage duplication.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant/cluster identifier |
| new_branch | string | yes | Name for the new branch (e.g. 'feature-migration') |
| parent_branch | string | no | Parent branch name (default: 'main') |
| target_lsn | number | no | Optional WAL LSN for point-in-time fork |
| mask_pii | boolean | no | Mask sensitive PII columns in child branch for GDPR/HIPAA compliance |
Promote a preview/ephemeral database branch to primary production root with atomic zero-downtime pointer swap.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| branch_id | string | yes | Branch ID to promote to primary root |
Fast-forward merge schema modifications and table additions from a staging/feature branch back to parent trunk in <200ms.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| branch_id | string | yes | Branch ID containing changes to merge |
| parent_branch | string | no | Parent target branch (default: 'main') |
Reset an ephemeral database branch back to its parent snapshot in <50ms, wiping all test mutations.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| branch_id | string | yes | Branch ID to reset |
Inspect SQL DDL schema diff and migration script between a branch and its parent.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| branch_id | string | yes | Branch ID to diff |
| parent_branch | string | no | Parent branch (default: 'main') |
Execute raw SQL query directly on a tenant database cluster or branch.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| query | string | yes | SQL query statement to execute |
Introspect database schema tables, column types, and primary key constraints.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
Configure Aurora Serverless v2 / Neon ACU scaling limits (min ACU, max ACU, scale-to-zero timeout).
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| min_acu | number | no | Minimum ACU capacity (e.g. 0.0 for scale-to-zero) |
| max_acu | number | no | Maximum ACU capacity (e.g. 16.0 or 128.0) |
| idle_timeout_sec | number | no | Inactivity seconds before scaling to zero (default: 300) |
Inspect or configure PgBouncer connection pooling mode and max connection limits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| pool_mode | string | no | Pooling mode |
| max_pool_size | number | no | Maximum pooled backend connections |
Manage 1-click PostgreSQL extensions (vector, pg_search, postgis, uuid-ossp).
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| action | string | yes | Action to perform |
| name | string | no | Extension name (for enable/disable) |
Inspect slow query execution diagnostics, call frequencies, and buffer cache hit rates.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
Read database branch state at a specific historical point in time (timestamp or LSN) via PostgreSQL continuous WAL replay.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| target_time | string | no | ISO 8601 target timestamp (e.g., 2026-11-15T01:00:00Z) |
| target_lsn | number | no | Numeric WAL Log Sequence Number |
| target_branch | string | no | Target branch name to instantiate (default: time-travel) |
Inspect the visual DAG ancestry and branch lineage tree of CoW database branches.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
Perform 3-way schema reconciliation and migration dry-run before merging an ephemeral branch into target.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| branch | string | yes | Source branch identifier to reconcile |
| target_branch | string | no | Target branch to reconcile against (default: main) |
Inspect slow query execution plans, automated FinOps index recommendations, and circuit breaker health.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
Create a persistent point-in-time storage snapshot of a database cluster branch.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| name | string | no | Optional snapshot identifier label |
| branch_id | string | no | Branch to snapshot (default: main) |
List all persistent snapshots taken on a database cluster.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
Retrieve continuous WAL archiving timelines and valid point-in-time recovery (PITR) restoration windows.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
Restore a database cluster to a historical LSN or timestamp on a new branch.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| target_branch | string | yes | Name of new branch for restored state |
| target_lsn | number | no | Target WAL Log Sequence Number |
Inspect PostgreSQL real-time Change Data Capture (CDC) streaming stats, buffer saturation, and subscriber lag.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
Publish or inject a transactional mutation event into the PostgreSQL real-time CDC event stream.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| table | string | yes | Target database table |
| action | string | yes | Mutation action |
| data | object | yes | Row mutation payload |
Initiate zero-downtime 1-click database migration/import from Neon, Supabase, or AWS RDS.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Destination database tenant identifier |
| source_url | string | yes | Source PostgreSQL connection URI |
Query current progress, replication lag, and dual-write replication status of active database import.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
Execute zero-downtime cutover, routing all connection pool traffic to Mnemon-DB.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
Configure pgvector + HNSW-PQ vector quantization with disk-spill indexing for sub-10ms billion-vector search.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
| table | string | yes | Target vector table |
| column | string | yes | Target vector embedding column |
| dimensions | number | no | Vector dimensions (default: 1536) |
| quantization_type | string | no | Quantization algorithm (default: PQ) |
| disk_spill | boolean | no | Enable disk-spill indexing for RAM conservation |
Inspect vector index memory savings, quantization ratios, and disk usage across pgvector tables.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| tenant_id | string | yes | Database tenant identifier |
Emergency termination of a rogue or long-running query by PID or duration threshold.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| pid | number | no | PostgreSQL backend process ID |
| query_id | string | no | Query identifier |
| duration_sec | number | no | Kill all queries running longer than this duration (seconds) |
Prune and reclaim storage extents from orphaned ephemeral sandbox or pull-request branches.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| max_age_hours | number | no | Reap branches older than this threshold in hours (default: 24) |
| dry_run | boolean | no | Dry run preview mode without actual extent deletion |
List persistent ext4 sparse volumes attached to a microVM sandbox for SQLite, PocketBase, or local file storage with scale-to-zero persistence.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sandbox_id | string | no | MicroVM sandbox ID (default: sbx-default) |
Attach a persistent sparse ext4 volume mounted at /data with scale-to-zero snapshot persistence.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sandbox_id | string | yes | MicroVM sandbox ID |
| mount_path | string | no | Target mount path inside microVM (default: /data) |
| size_gb | number | no | Virtual volume size in gigabytes (default: 10) |
Adaptive Dual-Liveness Probe: Probes HTTP endpoint and automatically falls back to raw TCP 3-way handshake on guest port, eliminating silent false-positive deploy failures.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| target_host | string | no | Target host IP or hostname (default: 127.0.0.1) |
| port | number | yes | Target port inside guest (e.g., 3000, 8080) |
| health_path | string | no | Health check URL path (e.g., /health, /) |
Zero-rebuild environment variable and secret injection via Firecracker MMDS in <50µs without rebuilding the SquashFS rootfs.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sandbox_id | string | yes | MicroVM sandbox ID |
| secrets | object | yes | Key-value map of secrets and environment variables to inject |
Autonomous CoW extent reaper: Reclaims unreferenced disk blocks and frees host RAM via process_madvise(MADV_DONTNEED).
No parameter schema on file.
Sub-35ms pgvector HNSW index clone: Duplicates vector index graphs across branches using XFS/Btrfs zero-copy reflink extents.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Mnemon project / tenant ID |
| branch | string | yes | Branch name to clone index from |
| index_name | string | yes | pgvector HNSW index name |
Configure automatic zero-CPU instance freezing via cgroups v2 (cgroup.freeze) while keeping PostgreSQL shared buffer pool in RAM.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Mnemon project / tenant ID |
| idle_timeout_seconds | number | no | Inactivity seconds before auto-freezing (default: 30) |
| preserve_buffer_pool | boolean | no | Preserve shared buffers in RAM (default: true) |
In-memory AST syntax and hazard detection (<50µs): Detects infinite loops, fork bombs, and banned syscalls before microVM dispatch.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| code | string | yes | Source code to analyze |
| language | string | yes | Programming language (python, javascript, c, cpp, bash, go) |
50-Testcase Parallel Fast-Batch: Dispatches up to 50 testcases in a single HTTP request to evaluate concurrently in 38ms parallel memory.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| language_id | number | no | Language ID (e.g. 54 for Python, 63 for JavaScript, 71 for Python 3) |
| source_code | string | yes | Source code to evaluate against tests |
| tests | array | no | List of up to 50 testcases with stdin and expected_output |
| submissions | array | no | Alternative batch submissions format for backward compatibility |
Parallel early-exit batch matrix execution: Runs test matrices across submissions concurrently with sub-millisecond short-circuiting on failure.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| language_id | number | no | — |
| source_code | string | no | — |
| tests | array | no | — |
| submissions | array | no | List of code submissions to evaluate |
Get real-time bidirectional WebSocket PTY terminal gateway connection details for an attached interactive microVM shell.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sandbox_id | string | yes | Target sandbox ID |
Allocate and expose a Layer 4 TCP wire forwarding port (antinat) from ephemeral host pool (12000-20000) for PostgreSQL, gRPC, or custom daemons.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sandbox_id | string | yes | Target sandbox ID |
| guest_port | number | yes | Internal port inside microVM (e.g. 5432, 8080) |
| protocol | string | no | Layer 4 transport protocol |
| service_name | string | no | Descriptive service name (e.g. postgres, grpc) |
List all active Layer 4 TCP port forwardings for a specific microVM sandbox.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sandbox_id | string | yes | Target sandbox ID |
Close an active Layer 4 TCP port forwarding and return the host port to the pool.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sandbox_id | string | yes | Target sandbox ID |
| host_port | number | yes | Allocated host port number to release |
Execute multi-candidate speculative branch racing: runs N candidate code patches in parallel ephemeral microVMs, commits the fastest passing candidate, and instantly reaps losers.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| problem_id | string | no | Unique identifier for the task or problem |
| candidates | array | yes | List of candidate hypotheses/patches to race in parallel |
| timeout_ms | number | no | Maximum race timeout in milliseconds |
31ms Point-in-Time Recovery & WAL rewind: forks an isolated zero-copy CoW PostgreSQL branch from an arbitrary historical timestamp or LSN.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sandbox_id | string | no | Optional sandbox ID to rewind |
| project_id | string | no | Optional project ID |
| target_time | string | no | Target ISO-8601 UTC timestamp to rewind to |
| target_lsn | string | no | Target PostgreSQL WAL LSN position |
List, add, or delete custom domains and verify SSL/TLS provisioning for a project.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Target project ID |
| action | string | yes | Action to perform |
| domain | string | no | Domain name (e.g., api.example.com), required for add and delete |
List, register, or delete scheduled cron jobs for serverless background invocations.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Target project ID |
| action | string | yes | Action to perform |
| path | string | no | HTTP route path to invoke (for add, e.g., /api/cron/cleanup) |
| schedule | string | no | Cron expression (for add, e.g., '0 * * * *') |
| cron_id | string | no | Cron job ID (required for delete) |
List, create, or delete Git / CI/CD deployment webhook triggers for autonomous build execution.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Target project ID |
| action | string | yes | Action to perform |
| name | string | no | Hook label (for create, e.g. GitHub Main Push) |
| branch | string | no | Target branch (for create, default: main) |
| hook_id | string | no | Hook ID (required for delete) |
List, set, or delete edge routing rules, rewrites, custom redirects, and HTTP security headers.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Target project ID |
| action | string | yes | Action to perform |
| source | string | no | Incoming path regex or pattern (for set, e.g. /api/(.*)) |
| destination | string | no | Target URL or rewrite path (for set, e.g. /$1) |
| type | string | no | Route rule type (default: rewrite) |
| status | number | no | HTTP redirect status code (e.g. 301, 302, 307, 308) |
| route_id | string | no | Route ID (required for delete) |
List, configure, or remove external SIEM log drains (Datadog, Papertrail, Axiom, HTTPS Webhooks).
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project_id | string | yes | Target project ID |
| action | string | yes | Action to perform |
| name | string | no | Drain name / label |
| provider | string | no | SIEM provider |
| url | string | no | HTTPS or syslog destination URL (required for add) |
| batch_size | number | no | Flush batch size (default: 100) |
| drain_id | string | no | Drain ID (required for delete) |
Inspect tenant credit wallet, balance (USD & credits), tier allocation, and compute boot limits.
No parameter schema on file.
Retrieve billing history, invoices, transaction receipts, and payment ledger entries.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| org_slug | string | no | Optional organization slug filter |
Inspect current authenticated identity, tenant ID, active tier, resource ceilings (RAM, vCPU, projects), and security quarantine status.
No parameter schema on file.
Execute code across any of the 10 Eschaton engine suites (Omni, Apex, Void, Complexity, Sanitizer, MemoryProfile, FileIO, DBMS, SpecialJudge) with custom testcase matrices.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| language_id | number | yes | Language ID (e.g. 54 for C++, 71 for Python 3, 202 for C++20, 204 for Python 3.11) |
| source_code | string | yes | Source code to execute |
| suite | string | no | Execution suite |
| test_cases | array | no | List of testcases with stdin, expected_output, and optional schema_sql or files |
| database | string | no | Database engine for DBMS suite |
| checker_code | string | no | Custom validator checker source code for SpecialJudge suite |
| checker_language_id | number | no | Checker language ID (default 202) |
| verbose | number | no | Verbose mode (0=early-exit short-circuit, 1=full diagnostics for Void suite) |
Deterministic cycle and instruction counter via Valgrind Lackey: measures exact guest instruction counts to mathematically prove O(N) vs O(N^2) complexity immune to CPU jitter.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| language_id | number | no | C or C++ language ID (e.g. 50, 54, 201, 202) |
| source_code | string | yes | C/C++ algorithm source code to profile |
| stdin | string | no | Standard input for algorithm |
Sub-20ms AddressSanitizer (ASan) + UndefinedBehaviorSanitizer (UBSan) safety check: detects buffer overflows, memory leaks, and use-after-free bugs.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| language_id | number | no | C or C++ language ID (e.g. 50, 54, 201, 202) |
| source_code | string | yes | C/C++ source code to sanitize |
| stdin | string | no | Standard input |
Automated SQL database grading across SQLite, PostgreSQL, MySQL, and Oracle: seeds an ephemeral database per testcase with schema_sql DDL/DML and evaluates queries.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| source_code | string | yes | Python database query/app code |
| schema_sql | string | yes | DDL/DML SQL statements to seed the test database |
| database | string | no | Database engine backend |
| expected_output | string | no | Expected query stdout |
| stdin | string | no | Optional standard input |
Isolated filesystem sandbox grading: seeds input files into sandbox disk, runs learner code that reads/writes files, and grades generated output file.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| language_id | number | no | Language ID |
| source_code | string | yes | Source code performing file I/O |
| input_filename | string | no | Input file name to seed |
| input_content | string | no | Content of input file |
| output_filename | string | no | Expected output file to read and grade |
| expected_output | string | no | Expected content of the output file |
Custom competitive programming checker binary validator: evaluates solutions with custom tolerances (e.g. 1e-6 float precision), multiple valid outputs, or graph permutations.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| language_id | number | no | Solution language ID |
| source_code | string | yes | Solution source code |
| checker_code | string | yes | Custom validator checker source code |
| checker_language_id | number | no | Checker language ID |
| stdin | string | no | Testcase stdin |
| expected_output | string | no | Testcase reference answer |
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under Apache-2.0
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install NEPHIS or to call it, at any volume.
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| verification-discount | repo | publisher verification (public source) — no provenance, but the source is public and inspectable | −1 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
Tools were statically extracted from the published source (75 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>".
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. high is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| thanaton_exec_command | code-exec | code-exec: keyword "exec" in tool name, parameter "command" |
| thanaton_run_code | code-exec | code-exec: keyword "run_code" in tool name |
| thanaton_spawn_sandbox | code-exec | code-exec: keyword "spawn" in tool name |
| thanaton_list_sandboxes | no tags | |
| thanaton_wake_sandbox | no tags | |
| thanaton_sleep_sandbox | no tags | |
| thanaton_destroy_sandbox | no tags | |
| thanaton_list_projects | no tags | |
| thanaton_create_project | no tags | |
| thanaton_deploy_code | no tags | |
| thanaton_get_logs | sensitive-source | sensitive-source: keyword "get_logs" |
| thanaton_manage_env | no tags | |
| thanaton_rollback | no tags | |
| thanaton_promote | no tags | |
| thanaton_configure_waf | no tags | |
| thanaton_configure_scaling | no tags | |
| mnemon_list_tenants | no tags | |
| mnemon_create_branch | no tags | |
| mnemon_promote_branch | no tags | |
| mnemon_merge_branch | no tags | |
| mnemon_reset_branch | no tags | |
| mnemon_diff_branch | no tags | |
| mnemon_execute_sql | sensitive-source | sensitive-source: keyword "execute_sql" |
| mnemon_get_schema | no tags | |
| mnemon_configure_scaling | no tags | |
| mnemon_configure_pooler | no tags | |
| mnemon_manage_extensions | no tags | |
| mnemon_get_slow_queries | no tags | |
| mnemon_time_travel | no tags | |
| mnemon_branch_lineage | no tags | |
| mnemon_reconcile_schema | no tags | |
| mnemon_query_advisor | no tags | |
| mnemon_create_snapshot | no tags | |
| mnemon_list_snapshots | no tags | |
| mnemon_pitr_timelines | no tags | |
| mnemon_pitr_restore | no tags | |
| mnemon_cdc_stats | no tags | |
| mnemon_cdc_publish | no tags | |
| mnemon_import_database | no tags | |
| mnemon_import_status | no tags | |
| mnemon_import_cutover | no tags | |
| mnemon_vector_quantize | no tags | |
| mnemon_vector_stats | no tags | |
| mnemon_admin_kill_query | no tags | |
| mnemon_admin_reap_branches | no tags | |
| thanaton_get_volumes | no tags | |
| thanaton_attach_volume | no tags | |
| thanaton_dual_probe | no tags | |
| thanaton_inject_secrets | no tags | |
| thanaton_reap_extents | no tags | |
| mnemon_clone_vector | no tags | |
| mnemon_autofreeze_policy | no tags | |
| eschaton_preflight_check | no tags | |
| nephis_fast_batch | external-sink | external-sink: keyword "http_request" |
| eschaton_fast_batch | no tags | |
| thanaton_pty_info | no tags | |
| thanaton_expose_port | no tags | |
| thanaton_list_ports | no tags | |
| thanaton_close_port | no tags | |
| thanaton_speculate_race | code-exec | code-exec: parameter "command" |
| mnemon_wal_rewind | no tags | |
| thanaton_manage_domains | no tags | |
| thanaton_manage_crons | no tags | |
| thanaton_manage_deploy_hooks | external-sink | external-sink: keyword "webhook" |
| thanaton_manage_routes | no tags | |
| thanaton_manage_log_drains | no tags | |
| thanaton_get_wallet | no tags | |
| thanaton_list_invoices | no tags | |
| nephis_whoami | no tags | |
| eschaton_execute_suite | sensitive-source | sensitive-source: parameter "sql" |
| eschaton_complexity_profile | no tags | |
| eschaton_sanitize_check | no tags | |
| eschaton_dbms_eval | code-execsensitive-source | code-exec: keyword "eval" in tool name; sensitive-source: parameter "sql" |
| eschaton_fileio_eval | code-exec | code-exec: keyword "eval" in tool name |
| eschaton_special_judge | no tags |
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
Supply chain and provenance
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 1.0.7. Pin a release below and the command asks for exactly that one.
No release note was published with this version.
- The version this listing was on when mcprush began following its releases. The registry was not asked when it shipped, so this row carries no date.
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Any signed-in account that does not publish NEPHIS can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.