Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/jpicklyk/task-orchestrator/deploy_to_dockergit clone --depth 1 https://github.com/jpicklyk/task-orchestratorWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00023 | $0.02932 |
| Opus 5 | $0.00012 | $0.01466 |
| Sonnet 5 | $0.00005 | $0.00586 |
| Haiku 4.5 | $0.00002 | $0.00293 |
Grade A, and why
deploy_to_docker scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 193 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Deploy to Docker
Build the MCP Task Orchestrator Docker image and start (or redeploy) the detached HTTP container that the active .mcp.json connects to. The Dockerfile handles its own Gradle build internally — no local build step needed.
On every deploy the command detects the current container's settings and defaults to reusing them, so you never have to re-specify things like API_AUTH_MODE=none after the first time. You can always review or change them via the chooser.
Usage: /deploy_to_docker [image-tag] [--build-only] [--run] [--reuse]
Default (no flags): Build the image, detect the existing container's config, then present a [Reuse / Reconfigure] chooser. Reuse redeploys with the same settings; Reconfigure walks the REST-mode / config-mount prompts. Matches the active .mcp.json (type: http, url: http://localhost:3001/mcp) — reachable after /mcp reconnect mcp-task-orchestrator.
Arguments:
image-tag(optional): Docker image tag. Defaulttask-orchestrator:current. A bare name likedev→task-orchestrator:dev.--build-only: Build the image only; start no container; no prompts.--run/--reconfigure: Build, then go straight to the reconfigure prompts (skip the reuse chooser) — including STDIO/HTTP transport choice.--reuse: Build, then redeploy with the detected settings without prompting (non-interactive redeploy).
Examples:
/deploy_to_docker— build, then Reuse-or-Reconfigure chooser/deploy_to_docker --reuse— rebuild + redeploy identical settings, no prompts/deploy_to_docker --run— build + full interactive reconfigure/deploy_to_docker --build-only— build image only
Workflow
Step 1: Resolve Image Tag
TARGET=runtime-current, default tagtask-orchestrator:current, data volumemcp-task-data.- A non-
--tag argument is used as-is; a tag with no:getstask-orchestrator:prepended; no arg → default.
Step 2: Build Docker Image
docker build --target runtime-current -t <image-tag> .
Multi-stage build compiles from source — no prior ./gradlew build.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 193 lines · 23 tokens per session scan A 60b1b4da4ff6
deploy_to_docker is a command published in the GitHub repository jpicklyk/task-orchestrator (205 stars, last pushed 28d ago), licensed MIT. It adds 23 tokens to every session and 2,932 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
paul:research-phase
Research unknowns for a phase using subagents.
paul:progress
Smart status with routing - suggests ONE next action.
paul:status
Display current loop position (PLAN/APPLY/UNIFY) and phase progress.
paul:unify
Reconcile plan vs actual and close the loop.
test
Smart test runner with filtering, coverage, and health monitoring.
safe-refactor
Safe refactoring with automated review, testing, and rollback capabilities.