NerveBLOCK
The Simple Agent Development Kit.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
The Simple Agent Development Kit
[](https://github.com/evilsocket/nerve/blob/main/docs/index.md) [](https://github.com/evilsocket/nerve/releases/latest) [](https://pypi.org/project/nerve-adk) [](https://hub.docker.com/r/evilsocket/nerve) [](https://github.com/evilsocket/nerve/actions/workflows/ci.yml) [](https://github.com/evilsocket/nerve/blob/master/LICENSE.md)
Join the project community on our server!
Nerve is a simple yet powerful Agent Development Kit (ADK) to build, run, evaluate, and orchestrate LLM-based agents using just YAML and a CLI. It’s designed for technical users who want programmable, auditable, and reproducible automation using large language models.
Key Features
📝 Declarative Agents
Defi
f3dbe779d24bOBSERVED · 2026-09-24Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add nerve-adk -- uvx nerve-adk
{
"mcpServers": {
"nerve-adk": {
"command": "uvx",
"args": [
"nerve-adk"
]
}
}
}Exposed tools (6)
6 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
calculate | read | A function that performs a calculation |
get_weather | read | Get the current weather in a given place. |
hello_world | read | A simple hello world function |
process_data | read | A function that processes complex data structures |
test_state_tool | read | A tool to test default string handling. |
tool_name | read | A test tool |
Trust audit
BLOCKgrade D · trust 63/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (5 observation(s))
- Network
- declared (4 observation(s))
- Shell
- declared (8 observation(s))
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (14)
exec(func_body, func_namespace)
exec(func_body, type_defs)
exec(code, func_namespace)
module = __import__(f"nerve.tools.namespaces.{modname}", fromlist=[""])module = __import__(f"nerve.tools.namespaces.{modname}", fromlist=[""])module = __import__(f"nerve.tools.namespaces.{namespace}", fromlist=[""]).pre-commit-config.yaml
module = __import__(f"nerve.tools.namespaces.{modname}", fromlist=[""])expected_result = eval(expression)
nerve run -g "ollama/llama3.2?api_base=http://1.2.3.4:11434" agent
api_base: http://127.0.0.1:8081
 and found 3 critical or high issues in the source. Each one is listed on this page with the file and line it is on.
What credentials does Nerve need?
No credential environment variables were found in its source, so it appears to need none.
How does Nerve run?
It speaks stdio and streamable-http, so it runs as a local process your client starts. It is published on PyPI as nerve-adk.
How current is this page?
The grade is for one exact copy of the source (f3dbe779d24b), read on 2026-09-24. The repository is watched and re-audited when it changes.