# pytest

Runs pytest and returns structured test results (passed, failed, errors, skipped, failures).

Agent View of the PolicyLayer registry record for `pytest`. HTML page: https://policylayer.com/tools/io-github-dave-london-pare-python/pytest

## Facts

- Tool: `pytest`
- Server: Python (`Dave-London/Pare`) — https://policylayer.com/tools/io-github-dave-london-pare-python.md
- Homepage: https://github.com/Dave-London/Pare
- Risk category: Execute (High risk)
- Registry record: grade F, identity unverified
- Server rate-limited: no
- Parameters: 12
- Recommended policy verdict: Rate-limited

## Parameters

| Parameter | Type | Required | Description |
| --- | --- | --- | --- |
| `env` | object | no | Additional environment variables for the pytest process, merged over the parent environment (e.g. {"PYTHONPATH": "src"} for src-layout projects) |
| `path` | string | no | Project root path |
| `compact` | boolean | no | Prefer compact output |
| `keyword` | string | no | Keyword expression for test filtering (-k EXPR) |
| `markers` | string | no | Pytest marker expression (e.g. "not slow") |
| `maxFail` | integer | no | Stop after N failures (--maxfail=N) |
| `targets` | array | no | Test files or directories to run (default: auto-discover) |
| `verbose` | boolean | no | Enable verbose output |
| `coverage` | string | no | Source directory for coverage measurement (--cov=SOURCE) |
| `parallel` | integer | no | Number of parallel workers for pytest-xdist (-n NUM, 0=auto) |
| `exitFirst` | boolean | no | Stop on first failure (-x) |
| `extraArgs` | array | no | Additional pytest CLI arguments passed through verbatim (e.g. ["-p", "no:logfire"] to disable a broken plugin). Each element is validated for control characters |

Parameters from the server's own tool schema.

## Example call (MCP tools/call, JSON-RPC 2.0)

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "pytest",
    "arguments": {}
  }
}
```

## Why pytest is rated High

Running pytest executes arbitrary Python code contained in test files. A malicious or misconfigured test suite could perform destructive operations, exfiltrate data, or cause side effects. The tool actively runs code rather than just reading or writing data, placing it firmly in Execute. Severity is high because the blast radius depends entirely on what the executed tests do, and there is no sandboxing implied.

From the tool's own definition: "'Runs pytest' — executes the pytest test runner, which runs arbitrary Python test code"

Risk signals: Accepts file system path (path) · High parameter count (18 properties)

## Use case

AI agents invoke pytest to trigger actions in Python. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.

## Recommended policy (PolicyLayer)

Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches Python:

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "pytest": {
      "limits": [
        {
          "counter": "pytest_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}
```

## Other tools on Python (13)

- `conda` — Execute — https://policylayer.com/tools/io-github-dave-london-pare-python/conda.md
- `mypy` — Execute — https://policylayer.com/tools/io-github-dave-london-pare-python/mypy.md
- `pip-install` — Execute — https://policylayer.com/tools/io-github-dave-london-pare-python/pip-install.md
- `poetry` — Execute — https://policylayer.com/tools/io-github-dave-london-pare-python/poetry.md
- `pyenv` — Execute — https://policylayer.com/tools/io-github-dave-london-pare-python/pyenv.md
- `ruff-check` — Execute — https://policylayer.com/tools/io-github-dave-london-pare-python/ruff-check.md
- `uv-install` — Execute — https://policylayer.com/tools/io-github-dave-london-pare-python/uv-install.md
- `uv-run` — Execute — https://policylayer.com/tools/io-github-dave-london-pare-python/uv-run.md
- `black` — Read — https://policylayer.com/tools/io-github-dave-london-pare-python/black.md
- `pip-audit` — Read — https://policylayer.com/tools/io-github-dave-london-pare-python/pip-audit.md
- `pip-list` — Read — https://policylayer.com/tools/io-github-dave-london-pare-python/pip-list.md
- `pip-show` — Read — https://policylayer.com/tools/io-github-dave-london-pare-python/pip-show.md
- `ruff-format` — Read — https://policylayer.com/tools/io-github-dave-london-pare-python/ruff-format.md

## For agents

This record is a snapshot. Live verdicts and the full registry:

- Check every server in your MCP config at once: `npx -y policylayer stack`
- Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md)
- Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events`

---

Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=io-github-dave-london-pare-python · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/io-github-dave-london-pare-python