Skip to content
Agent Detective

Installation and deployment paths

Installation and deployment paths

Section titled “Installation and deployment paths”

Choose how you run agent-detective on a host or in your cluster. This page is the entry point; the detailed guides are linked below.

Typical reading order: this page (choose how you run) → configuration hub (how settings load) → upgrading (releases and image tags).

Choose a deployment style

Section titled “Choose a deployment style”
PathWhen to useWhat you need
Container image (GHCR)Production on a single host or small team; minimal build stepsDocker (or compatible runtime), a config/ directory to mount, secrets via env (see configuration.md)
Docker Compose (build or pull)Same as above, but you want compose and optional bind mounts for config/ and plugins/docker.md (includes docker-compose.ghcr.yml for pull-only)
From source (git + pnpm)You fork the repo, change packages/, or run without a prebuilt imageNode.js 24+, pnpm 10+ (see root package.json packageManager), git
Bare metal: systemd + reverse proxyNo Docker; long-running service on a VM with nginx or similardeployment.md (systemd, nginx, sizing)

Kubernetes or Helm: this repository does not ship charts. Run the GHCR image with your platform’s standard workload manifest and the same config + env model as in docker.md.

Host capabilities

Section titled “Host capabilities”
  • Process: either the container image (includes a bundled node + built app) or Node.js 24+ when building from source.
  • Configuration: JSON under config/ (configuration.md); optional config/local.json (often gitignored) for secrets and overrides.
  • Repositories: the local-repos plugin needs git and filesystem access to the repos you list in config (bind mounts in Docker, or local paths on bare metal).
  • Network: outbound to Jira (if you use the Jira plugin), to git remotes (for pr-pipeline), and to your AI provider as required by the agent CLI (e.g. OpenCode). Inbound: HTTP(S) to the app (webhooks, API).
  • Agent CLIs: the image or host must be able to run the configured agent (e.g. opencode in the default image). See docker.md and cursor-agent.md for adding other agents.

Configuration (all paths)

Section titled “Configuration (all paths)”
  1. Read the configuration hub for load order and top-level config shape.
  2. Use configuration.md for the full env whitelist and plugin narratives.
  3. Use config/local.json and/or the env whitelist for secrets in production.
  4. For plugin option fields, use generated/plugin-options.md.

Detailed guides

Section titled “Detailed guides”
TopicDocument
Docker, Compose, production image, GHCRdocker.md
systemd, nginx, health checks, troubleshooting (no Docker)deployment.md
Config files, env, pluginsconfiguration.md
Releases, pinning images, git upgradeupgrading.md
Jira E2E (tunnel, webhooks, pr-pipeline)e2e/README.md
Day-to-day monorepo developmentdevelopment.md

Clone URL (from source)

Section titled “Clone URL (from source)”

If you build from a clone, use the upstream repository (or your fork’s URL):

Terminal window
git clone https://github.com/toniop99/agent-detective.git
cd agent-detective

Replace toniop99/agent-detective with your fork’s owner/name on GitHub if applicable.

See also

Section titled “See also”