npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@infersec/conduit

v1.117.0

Published

End user conduit agent for connecting local LLMs to the cloud.

Downloads

2,507

Readme

@infersec/conduit

Conduit agent for connecting local LLMs to the Infersec cloud.

Usage

npx @infersec/conduit inference start --engine <type> --key <api-key> --source <source-id>

Commands

| Command | Description | | ----------------------- | ------------------------------------------------------------------------------------------------------ | | inference start | Start the inference agent | | inference group start | Start an inference agent for a source group (registers an ephemeral source) | | engine create | Create or update an engine resource | | models create | Create or update a registered model resource | | models list | List cached models and disk usage | | models clear | Remove cached models | | models fit | Detect local hardware and suggest models that will run | | source create | Create or update an inference source resource | | endpoint create | Create or update an inference endpoint resource | | tool connect <toolID> | Connect a local tool (local-filesystem, local-mysql) to the API. Type is resolved from the server. |

Note: The resource management commands (engine create, models create, source create, endpoint create) and models fit have no environment variable fallbacks - all options are provided as flags. The inference and tool commands keep their env var equivalents.

Flags

| Flag | Required | Default | Notes | | --------------- | -------- | ------------------------------ | -------------------------------------------------------------------------------------- | | --engine | Yes | - | Engine type (matches ENGINE). | | --engine-port | No | 9700 | Engine port (matches ENGINE_PORT). | | --key | Yes | - | API key (matches API_KEY). | | --port | No | 9600 | Port to listen on (matches PORT). | | --root | No | $HOME/.cache/infersec/iagent | Root directory (matches ROOT_DIRECTORY). | | --source | Yes | - | Inference source ID (matches SOURCE). | | --api-url | No | https://api.infersec.ai | API base URL (matches API_URL). | | --start-mode | No | auto | Startup mode (matches START_MODE): auto starts engine, idle leaves conduit idle. |

inference group start flags

Same flags as inference start, except --group replaces --source:

| Flag | Required | Default | Notes | | --------- | -------- | ------- | ------------------------------------- | | --group | Yes | - | Source group ID (matches GROUP_ID). |

The agent registers an ephemeral source under the group (absorbing its model/engine settings), then runs the standard inference flow. Ephemeral sources are removed automatically after being offline for 1 hour.

Resource management commands

engine create, models create, source create and endpoint create manage Infersec resources via the public REST API. They all share these flags (no environment variable fallbacks):

| Flag | Required | Default | Notes | | ----------- | --------------- | ------- | ------------------------------------------------------------- | | --api-url | Yes | - | API base URL. | | --key | Yes | - | API key. | | --name | Yes | - | Resource name (used for --update matching). | | --update | No | false | Update an existing resource matched by --name, else create. | | --id | With --update | - | Target an existing resource by ID instead of name matching. |

Per-command flags:

| Command | Flags | | ----------------- | --------------------------------------------------------------------------------------------------------------------------- | | engine create | --type <llama.cpp\|vllm\|sglang\|tensorrt-llm\|mlx-lm\|exllamav3> (required), --arg <flag> (repeatable raw engine flag) | | models create | --slug <owner/repo> (required), --format <gguf\|safetensors\|...> (required) | | source create | --engine <id> (required), --model <id> (required), --quant <label>, --context-length <n> | | endpoint create | --source <id> (repeatable, at least one), --group <id> (repeatable), --routing-method, --enabled |

models fit

Detects local hardware (GPU compute, memory, engines including Docker), downloads a small probe model (LFM2.5-350M, served from the Infersec asset mirror - LFM Open License v1.0) and suggests models that fit. Prints a runnable command chain to deploy the top suggestion.

Backend resolution order: local llama.cpp (GGUF), local vLLM (safetensors), then Docker vLLM images when Docker is available. Without any backend, falls back to rule-based suggestions.

| Flag | Required | Default | Notes | | --------------- | -------------------- | ------------------------------ | ---------------------------------------------- | | --api-url | No | https://api.infersec.ai | API used to resolve probe model download URLs. | | --interactive | No | - | Agent session that deploys suggestions. | | --key | With --interactive | - | API key for the interactive session. | | --json | No | - | Machine-readable JSON output. | | --probe-model | No | LiquidAI/LFM2.5-350M | Probe model slug requested from the API. | | --root | No | $HOME/.cache/infersec/iagent | Root directory for probe model files. |

tool connect flags

Common flags (all tool types):

| Flag | Required | Default | Notes | | ----------- | -------- | ------------------------- | --------------------------------------- | | <toolID> | Yes | - | Tool ID to connect (matches TOOL_ID). | | --key | Yes | - | API key (matches API_KEY). | | --api-url | No | https://api.infersec.ai | API base URL (matches API_URL). |

local-filesystem flags:

| Flag | Required | Default | Notes | | ------------- | -------- | ----------------- | ----------------------------------------------------------------------- | | --path | No | . (current dir) | Filesystem base path (matches TOOL_PATH). Paths are confined to this. | | --read-only | No | false | Expose only read operations (matches READ_ONLY). |

local-mysql flags (MySQL/MariaDB). Defaults to read-only; use --allow to enable writes:

| Flag | Required | Default | Notes | | -------------------- | --------------- | -------- | ------------------------------------------------------------------------------------------------------ | | --url | One of url/host | - | Connection URL mysql://user:pass@host:3306/db (matches MYSQL_URL). Prefer MYSQL_URL for secrets. | | --host | One of url/host | - | Host (matches MYSQL_HOST). | | --port | No | 3306 | Port (matches MYSQL_PORT). | | --user | With host | - | User (matches MYSQL_USER). | | --password | With host | - | Password (matches MYSQL_PASSWORD). Prefer MYSQL_PASSWORD to avoid shell history exposure. | | --database | No | - | Database name (matches MYSQL_DATABASE). | | --allow | No | (none) | Write operations to enable (csv) or all (matches MYSQL_ALLOW). Read-only if unset. | | --pool-max | No | 5 | Pool connection limit (matches MYSQL_POOL_MAX). | | --max-rows | No | 1000 | Max rows returned per query (matches MYSQL_MAX_ROWS). | | --query-timeout-ms | No | 30000 | Per-query timeout in ms (matches MYSQL_QUERY_TIMEOUT_MS). |

Examples

| Scenario | Command | | ------------------ | --------------------------------------------------------------------------------------------------------------------------------- | | Required only | npx @infersec/conduit inference start --engine vllm --key <api-key> --source <source-id> | | Custom root/port | npx @infersec/conduit inference start --engine vllm --key <api-key> --source <source-id> --root /data/infersec --port 9601 | | Custom engine/port | npx @infersec/conduit inference start --engine vllm --key <api-key> --source <source-id> --port 9601 --engine-port 9701 | | Source group | npx @infersec/conduit inference group start --key <api-key> --group <group-id> | | Connect FS tool | npx @infersec/conduit tool connect <tool-id> --key <api-key> --path ./ | | Read-only FS tool | npx @infersec/conduit tool connect <tool-id> --key <api-key> --path ./ --read-only | | Connect MySQL tool | MYSQL_URL=mysql://user:pass@host:3306/db npx @infersec/conduit tool connect <tool-id> --key <api-key> | | MySQL + writes | MYSQL_URL=mysql://user:pass@host:3306/db npx @infersec/conduit tool connect <tool-id> --key <api-key> --allow insert,update,sql |

Credentials are best supplied via the MYSQL_* environment variables (or MYSQL_URL) rather than --password/--url flags, to avoid exposing them in process listings and shell history.

Environment variables

| Variable | Required | Default | Notes | | ---------------- | -------- | ------------------------------ | ---------------------------------------------------------------------------- | | AUTO_PORTS | No | false | Auto-allocate conduit and engine ports. Conflicts with PORT/ENGINE_PORT. | | ENGINE | Yes | - | Engine type (matches --engine). | | ENGINE_PORT | No | 9700 | Engine port (matches --engine-port). | | API_KEY | Yes | - | API key (matches --key). | | GROUP_ID | No | - | Source group ID for inference group start (matches --group). | | SOURCE | Yes | - | Inference source ID (matches --source). | | API_URL | No | https://api.infersec.ai | API base URL (matches --api-url). | | PORT | No | 9600 | Port to listen on (matches --port). | | ROOT_DIRECTORY | No | $HOME/.cache/infersec/iagent | Root directory (matches --root). | | START_MODE | No | auto | Startup mode (matches --start-mode). |

tool connect environment variables

Common:

| Variable | Required | Default | Notes | | --------- | -------- | ------------------------- | --------------------------------------------------- | | API_KEY | Yes | - | API key (matches --key). | | TOOL_ID | Yes | - | Tool ID to connect (matches <toolID> positional). | | API_URL | No | https://api.infersec.ai | API base URL (matches --api-url). |

local-filesystem:

| Variable | Required | Default | Notes | | ----------- | -------- | ------- | ---------------------------------------------------- | | TOOL_PATH | No | . | Filesystem base path (matches --path). | | READ_ONLY | No | false | Expose only read operations (matches --read-only). |

local-mysql:

| Variable | Required | Default | Notes | | ------------------------ | --------------- | -------- | ------------------------------------------------------- | | MYSQL_URL | One of url/host | - | Connection URL (matches --url). | | MYSQL_HOST | One of url/host | - | Host (matches --host). | | MYSQL_PORT | No | 3306 | Port (matches --port). | | MYSQL_USER | With host | - | User (matches --user). | | MYSQL_PASSWORD | With host | - | Password (matches --password). | | MYSQL_DATABASE | No | - | Database name (matches --database). | | MYSQL_ALLOW | No | (none) | Write ops csv / all (matches --allow). | | MYSQL_POOL_MAX | No | 5 | Pool connection limit (matches --pool-max). | | MYSQL_MAX_ROWS | No | 1000 | Max rows per query (matches --max-rows). | | MYSQL_QUERY_TIMEOUT_MS | No | 30000 | Per-query timeout in ms (matches --query-timeout-ms). |

CLI flags override environment variables when both are provided.