# agentl (pypi · agentl-etl)

Validate, plan, preview and run YAML data pipelines (ETL with LLM steps) and debug failed runs.

- Trust score: 79/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-10-19

## Components

- pypi · `agentl-etl`: 79/100 (this document), [markdown](https://verifymcp.io/servers/arimayuji-agentl/agentl-etl.md), [page](https://verifymcp.io/servers/arimayuji-agentl/agentl-etl)

## Channel facts

- Registry: `pypi`
- Package: `agentl-etl`
- Version: `0.9.0`
- Transport: `stdio`

## Trust breakdown

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. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-10-19.

- **Supply Chain Security**: 100/100
  - No malware found by supply-chain analysis.
  - No known CVEs affecting this package version or its production dependencies.
  - Runs hatchling.build at install time, a recognised build step with no custom scripting around it.
  - 0 of 6 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 100/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to arimayuji/agentl).
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 1 days ago).
  - Publishes a security disclosure policy (SECURITY.md).
- **Schema Quality & AI Usability**: 65/100
  - AI-judged instruction clarity (good).
  - Tool/resource definitions use about 580 tokens (~44/item across 13 items; 13 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability not yet verified: not enough scan history yet (needs a 30-day window).
- **Tool Coverage**: 90/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 69% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 13 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 13 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

**Unverified: 1 category.** A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.

## Install

### How do I install the agentl MCP server?

agentl runs locally as a PyPI package, launched with uvx agentl-etl. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add arimayuji-agentl -- uvx agentl-etl
```

### Cursor

```json
{
  "mcpServers": {
    "arimayuji-agentl": {
      "command": "uvx",
      "args": [
        "agentl-etl"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "arimayuji-agentl": {
      "command": "uvx",
      "args": [
        "agentl-etl"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add arimayuji-agentl -- uvx agentl-etl
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "arimayuji-agentl": {
      "type": "local",
      "command": [
        "uvx",
        "agentl-etl"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add arimayuji-agentl --command uvx --arg agentl-etl
```

### Hermes

```yaml
mcp_servers:
  arimayuji-agentl:
    command: "uvx"
    args: ["agentl-etl"]
```

### Netclaw

```json
{
  "McpServers": {
    "arimayuji-agentl": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "agentl-etl"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add arimayuji-agentl -t stdio -c uvx -a agentl-etl
```

### Other

```json
{
  "mcpServers": {
    "arimayuji-agentl": {
      "command": "uvx",
      "args": [
        "agentl-etl"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-10-16 (score 79)

First indexed and scored.

## MCP tools (13)

### `validate` (~33 tokens)

Check a pipeline without running it.

Input parameters:

- `pipeline` (string, required): path to a pipeline YAML file, or the YAML text itself

### `diagram` (~33 tokens)

The pipeline as a Mermaid diagram.

Input parameters:

- `pipeline` (string, required): path to a pipeline YAML file, or the YAML text itself

### `lint` (~58 tokens)

Validate plus best-practice warnings (missing budget, delivery with no check, secrets in the YAML, unused steps). JSON findings with line, code and fix.

Input parameters:

- `pipeline` (string, required): path to a pipeline YAML file, or the YAML text itself

### `plan` (~71 tokens)

What a run would read, call and cost: records per step, cache hits, LLM calls, token and cost ranges, and what would be written. Reads the sources; never writes, never calls an LLM, never runs fn: code.

Input parameters:

- `pipeline` (string, required): path to a pipeline YAML file

### `preview_step` (~70 tokens)

The first records going into and out of one step, computed like plan (readers and transforms only). Use it to check a transform before running the pipeline.

Input parameters:

- `n` (integer): default 5
- `pipeline` (string, required): path to a pipeline YAML file
- `step` (string, required)

### `new_pipeline` (~58 tokens)

Write a validated pipeline YAML from a description (calls the LLM set by AGENTL_PROVIDER / AGENTL_MODEL, which costs a little). Returns the YAML; nothing is saved.

Input parameters:

- `description` (string, required): what the pipeline should do

### `run_pipeline` (~78 tokens)

Run a pipeline file and return its status, spend and how many records each step produced. Pipelines that run Python code (fn: steps, Python tools, local MCP servers) need the server started with --allow-code.

Input parameters:

- `no_cache` (boolean): ignore cached agent answers
- `pipeline` (string, required): path to a pipeline YAML file

### `list_runs` (~18 tokens)

Recent runs.

Input parameters:

- `limit` (integer)

### `get_run` (~27 tokens)

A run with its events (the last 200).

Input parameters:

- `run_id` (string, required)

### `list_failures` (~29 tokens)

Recent failed runs, each with the failing step and its error.

Input parameters:

- `limit` (integer)

### `pipeline_stats` (~69 tokens)

Track record per pipeline: success rate, durations, cost, failures grouped by reason and a per-day trend.

Input parameters:

- `pipeline` (string): pipeline name; all when omitted
- `since` (string): how far back, e.g. 7d, 12h (default 30d)

### `doctor` (~21 tokens)

Explain why a run failed.

Input parameters:

- `run_id` (string, required)

### `list_connectors` (~15 tokens)

Installed connectors and their parameters.

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Dependencies. The full working is on the page: https://verifymcp.io/servers/arimayuji-agentl/agentl-etl#diagnostics

## Score history

- 2026-10-19: 79
- 2026-10-18: 79
- 2026-10-17: 79
- 2026-10-16: 79

## Common questions

### What is the agentl MCP server?

agentl is an MCP server listed in the public MCP registry as io.github.arimayuji/agentl. Validate, plan, preview and run YAML data pipelines (ETL with LLM steps) and debug failed runs. This page covers its PyPI package (agentl-etl).

### Is the agentl MCP server safe to use?

agentl scores 79 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 19 October 2026. Its build provenance is signed and verified. 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 agentl MCP server expose?

agentl exposes 13 tools: validate, diagram, lint, plan, preview_step, and 8 more. Their descriptions and schemas cost roughly 580 tokens of context every time the server is loaded.

### Is the agentl MCP server still maintained?

agentl is still listed as active in the MCP registry. We last reached this channel on 19 October 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 agentl MCP server under?

agentl declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- PyPI project: https://pypi.org/project/agentl-etl/
- Socket report: https://socket.dev/pypi/package/agentl-etl
- Repository: https://github.com/arimayuji/agentl
- Website: https://agentl.com.br/
- Changelog RSS feed: https://verifymcp.io/servers/arimayuji-agentl/agentl-etl.xml
- Changelog JSON feed: https://verifymcp.io/servers/arimayuji-agentl/agentl-etl.json
- HTML version of this page: https://verifymcp.io/servers/arimayuji-agentl/agentl-etl