freellmpool
PYPI · FREELLMPOOL · SCANNED SEP 20
OpenAI-compatible MCP gateway cataloging 22 LLM providers, with routing, failover, and tools.
Available components
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 Security100
- No malware found by supply-chain analysis.Pass
- No known CVEs affecting this package version or its production dependencies.Pass
- Runs hatchling.build at install time, a recognised native-build step with no shell scripting around it. View diagnostics → Pass
- 0 of 9 dependencies flagged as unhealthy. View diagnostics → Pass
Provenance & Transparency35
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- License check failed: the license (MIT License) isn't a recognized OSI-approved license. See how to fix → Fail
- Actively maintained (last published 20 days ago).Pass
- Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability71
- AI-judged instruction clarity (excellent).Pass
- Context-footprint check failed: tool/resource definitions use about 1963 tokens (~151/item across 13 items; 13 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 Management93
- Stability observed for 28 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage100
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 100% of tool parameters carry a description.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 13 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 14 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
How do I install the freellmpool MCP server?
freellmpool runs locally as a PyPI package, launched with uvx freellmpool. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
pypi · freellmpool
claude mcp add 0xzr-freellmpool -- uvx freellmpool
{
"mcpServers": {
"0xzr-freellmpool": {
"command": "uvx",
"args": [
"freellmpool"
]
}
}
} {
"servers": {
"0xzr-freellmpool": {
"command": "uvx",
"args": [
"freellmpool"
]
}
}
} codex mcp add 0xzr-freellmpool -- uvx freellmpool
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"0xzr-freellmpool": {
"type": "local",
"command": [
"uvx",
"freellmpool"
],
"enabled": true
}
}
} openclaw mcp add 0xzr-freellmpool --command uvx --arg freellmpool
mcp_servers:
0xzr-freellmpool:
command: "uvx"
args: ["freellmpool"] {
"McpServers": {
"0xzr-freellmpool": {
"Transport": "stdio",
"Command": "uvx",
"Arguments": [
"freellmpool"
]
}
}
} assistant mcp add 0xzr-freellmpool -t stdio -c uvx -a freellmpool
{
"mcpServers": {
"0xzr-freellmpool": {
"command": "uvx",
"args": [
"freellmpool"
]
}
}
} 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.
- 20 Sept 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.
- 18 Sept 26 −2
- Stability: pass → 0.87 functional
- 17 Sept 26 +1
- Stability: 0.97 → pass security
- 15 Sept 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.
- 13 Sept 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.
- 11 Sept 26 −3
- Stability: pass → 0.80 functional
- 10 Sept 26 +1
- Stability: 0.97 → pass security
- 8 Sept 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.
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 20 Sept 2026 · Analysed pypi/freellmpool@0.13.0
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | pypi |
Background: How many MCP packages publish verified provenance →
Install scripts 1 script
| Hook | Tier | Command |
|---|---|---|
| build_backend | allowlisted | hatchling.build |
Background: Why install scripts are a supply-chain risk →
Dependencies 9 packages
| Packages resolved | 9 |
|---|---|
| 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 →
free_llm_ask ~232
Ask a free LLM (pooled across configured free providers, with automatic failover). Offload a self-contained subtask — drafting, summarizing, classifying, brainstorming, a quick lookup — to a free model. The reply tells you which provider/model actually served it.
| Name | Type | Req | Description |
|---|---|---|---|
| max_tokens | integer | – | Max output tokens (default 1024). |
| model | string | – | Optional model name or provider/model (e.g. groq/llama-3.3-70b-versatile). Default: auto. |
| prompt | string | yes | The question or task. |
| provider | string | – | Optional provider id to restrict to (e.g. groq, cerebras). |
| routing | string | – | How to pick the model: agent (strongest healthy tier with quota spreading), quality (capability matched to prompt), spread (whole-pool breadth), fast (lowest latency), fair (provider balance), or aut… |
| system | string | – | Optional system instruction. |
| task | string | – | Optional task hint for quality routing; auto classifies locally. |
No output schema declared.
No examples provided.
free_llm_battle ~166
Compare a prompt across a small bounded panel of free models and render the result as a Markdown comparison table. Bounded to a few models (2-5) — for the broader eligible-target stress test use `tokenmax`. Per-model failures stay visible in the rendered output rather than failing the whole call.
| Name | Type | Req | Description |
|---|---|---|---|
| max_tokens | integer | – | Max output tokens per model (default 512). |
| n | integer | – | How many distinct models to ask (2-5, default 3). |
| prompt | string | yes | The question or task to compare across models. |
| routing | string | – | How to rank candidate panel models (default: quality). |
| synthesize | boolean | – | If true, a quality-routed model synthesizes the panel into one best answer. |
No output schema declared.
No examples provided.
free_llm_models ~25
List the available free provider/model ids freellmpool can route to.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
free_llm_panel ~172
Ask the SAME prompt to several different free models at once and get every answer back side by side - the agent-facing second-opinion surface. Great for cross-checking a fact, comparing approaches, or reducing single-model bias. Optionally have a strong model synthesize the best combined answer.
| Name | Type | Req | Description |
|---|---|---|---|
| max_tokens | integer | – | Max output tokens per model (default 512). |
| n | integer | – | How many distinct models to ask (2-5, default 3). |
| prompt | string | yes | The question or task to ask every model. |
| routing | string | – | How to rank candidate panel models (default: quality). |
| synthesize | boolean | – | If true, a quality-routed model synthesizes the panel into one best answer. |
| system | string | – | Optional system instruction. |
No output schema declared.
No examples provided.
free_llm_quota ~37
Show today's free-tier usage (UTC): per-provider request counts and daily-limit headroom, plus session totals and estimated cost avoided.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
free_llm_quota_wise ~79
Show local quota-wise status and headroom advice built from your locally tracked counters. Active mode, recommended mode, and per-provider used / limit / remaining. Output NEVER recommends account rotation or rate-limit bypass — it only suggests waiting for reset, lowering fan-out or token budget, or making an explicit paid choice outside the default flow.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
free_llm_recipe ~260
Run a bundled recipe end-to-end (e.g. `pr-review`, `second-opinion`, `repo-summary`). Recipes are bounded, role-driven, and pre-shaped — supply `name` plus whatever inputs the recipe declares. Missing recipe input or template variables return a tool error, never a traceback.
| Name | Type | Req | Description |
|---|---|---|---|
| input | string | – | Optional alias for `prompt` (recipe `input` template variable). |
| max_tokens | integer | – | Max output tokens (default 1024 for text recipes; panel recipes honor recipe defaults). |
| name | string | yes | Recipe name (e.g. `pr-review`, `second-opinion`, `repo-summary`). |
| opinions | integer | – | Panel size for panel-output recipes (2-5, default 3). |
| path | string | – | Optional glob for path-input recipes (e.g. `repo-summary --path 'src/**/*.py'`). |
| prompt | string | – | Optional inline text input (used for text-input recipes). |
| synthesize | boolean | – | If true, a quality-routed model synthesizes the panel into one best answer. |
| validation_output | string | – | Optional validation/test output for recipes that template it (e.g. `metaswarm-worker-review`). |
No output schema declared.
No examples provided.
free_llm_roles ~80
List bundled ask-role presets (e.g. `coder`, `critic`, `summarizer`, `second-opinion`) with their routing mode, max-tokens, temperature, and recommended use. Pass `name` to get the full details for one role.
| Name | Type | Req | Description |
|---|---|---|---|
| name | string | – | Optional role name to fetch a single role's details. |
No output schema declared.
No examples provided.
free_llm_route ~93
Explain where a prompt WOULD be routed without spending a single token: the estimated difficulty and the ranked list of candidate models (with capability scores) for the chosen routing mode. Use it to understand or debug routing.
| Name | Type | Req | Description |
|---|---|---|---|
| prompt | string | yes | The prompt to analyze. |
| routing | string | – | Routing mode to explain (default: the server's mode). |
| task | string | – | Optional task hint to explain. |
No output schema declared.
No examples provided.
free_llm_second_opinion ~182
Run the shared small-panel second-opinion flow. Same behavior as `free_llm_panel` — exposed as its own agent-facing tool so callers can declare their intent (a *second opinion*) without reasoning about the panel primitive. The panel size is bounded (2-5) and the per-model token budget is clamped.
| Name | Type | Req | Description |
|---|---|---|---|
| max_tokens | integer | – | Max output tokens per model (default 512). |
| n | integer | – | How many distinct models to ask (2-5, default 3). |
| prompt | string | yes | The question or task to ask every model. |
| routing | string | – | How to rank candidate panel models (default: quality). |
| synthesize | boolean | – | If true, a quality-routed model synthesizes the panel into one best answer. |
| system | string | – | Optional system instruction. |
No output schema declared.
No examples provided.
free_llm_stats ~52
Show freellmpool's LIFETIME totals (persisted across restarts): tokens served free, requests, and estimated cost avoided vs Claude Opus 4.8 — the number that keeps growing.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
free_llm_tailnet_info ~112
Show safe Tailscale Tailnet connection instructions for serving `freellmpool proxy` on another machine. Includes the local Tailnet IPv4, status, and OpenAI / Anthropic client env-var hints. Output NEVER contains a real local bearer token — it uses a `<proxy-key>` placeholder — and never leaks provider API keys. Degrades cleanly when `tailscale` is absent.
| Name | Type | Req | Description |
|---|---|---|---|
| port | integer | – | Optional proxy port to include in setup hints (default 8080). |
No output schema declared.
No examples provided.
tokenmax ~227
🌈 TOKENMAX 🌈 — gloriously excessive: fan the SAME prompt out to automatically eligible ranked targets across configured providers (hard cap 256; a max, routing policy, or CLI wise mode may narrow the swarm), then YOU (the calling model) synthesize the single best answer from the returned responses. Maximum cross-checking. Tongue-in-cheek, but genuinely useful for hard questions. Call this tool DIRECTLY (your client receives live `🌈 TOKENMAXXING ▸ N/total` progress as each model answers) — do NOT shell out to the CLI, which hides that from the user. To let the human watch the flashing rainbow, suggest they run `freellmpool tokenmax "<prompt>"` in their own terminal.
| Name | Type | Req | Description |
|---|---|---|---|
| max_models | integer | – | Optional cap that narrows the eligible ranked set (hard max 256). |
| max_tokens | integer | – | Max output tokens per model (default 400). |
| prompt | string | yes | The prompt to send to the selected eligible targets. |
| system | string | – | Optional system instruction. |
No output schema declared.
No examples provided.
What is the freellmpool MCP server?
freellmpool is an MCP server listed in the public MCP registry as io.github.0xzr/freellmpool. OpenAI-compatible MCP gateway cataloging 22 LLM providers, with routing, failover, and tools. This page covers its PyPI package (freellmpool).
Is the freellmpool MCP server safe to use?
freellmpool scores 81 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 20 September 2026. 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 freellmpool MCP server expose?
freellmpool exposes 13 tools: free_llm_ask, free_llm_panel, free_llm_second_opinion, free_llm_battle, free_llm_recipe, and 8 more. Their descriptions and schemas cost roughly 1,717 tokens of context every time the server is loaded.
Is the freellmpool MCP server still maintained?
freellmpool is still listed as active in the MCP registry. We last reached this channel on 20 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.