worker-start

A command that starts a Worker session for one of three disciplines: TDD, review, or validation. TDD means writing tests to guide the implementation before or alongside code changes.

In plain words
What is it for?
Use it to start a worker focused on test-driven development, code review, or validation of existing work.
Why use it?
It tells the session which kind of engineering work to look for and lets the stop hook dispatch matching board items automatically. It also rejects missing or unsupported discipline values.

Command

Install

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.

agentmods
npx agentmods add commands/ghostlygawd/engineering-board/worker-start
Clone the repo
git clone --depth 1 https://github.com/GhostlyGawd/engineering-board
Per session 53 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,419 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.00053 $0.01419
Opus 5 $0.00026 $0.00709
Sonnet 5 $0.00011 $0.00284
Haiku 4.5 $0.00005 $0.00142

Measured yesterday against content hash 37e37c681882, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

worker-start 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 yesterday.

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.

commands/worker-start.md · 103 lines

How it starts

The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/worker-start: start Worker mode

Mark the current session as an engineering-board Worker session for a specific discipline. While session-mode=worker, the Stop hook runs the Worker continuation procedure: search the live board for needs:<discipline> entries, claim one via the atomic mkdir primitive, dispatch the matching worker subagent, release the claim, and emit a continuation sentinel.

What to do

You are writing a small JSON state file that the Stop hook reads. Be precise about path, shape, and discipline validation.

Step 1: Parse the --discipline argument

The argument list is $ARGUMENTS. Look for --discipline <value> (one-token value, kebab-case). Accept the long form --discipline=value as well.

  • If --discipline is missing, print Usage: /worker-start --discipline <tdd|review|validate>. No action taken. and stop.
  • If <value> is empty, print the same usage message and stop.
  • If <value> is anything other than tdd, review, or validate, print Engineering board: unsupported discipline "<value>". supported disciplines: tdd, review, validate. No action taken. and stop.

The supported-discipline set for this milestone is exactly {"tdd","review","validate"}. Do not silently coerce or accept aliases.

Step 2: Resolve paths

  • State directory: ${CLAUDE_PROJECT_DIR}/.engineering-board/
  • State file: ${CLAUDE_PROJECT_DIR}/.engineering-board/session-mode.json

Create the state directory if it does not exist (mkdir -p semantics).

Step 3: Run the mode-transition guard

Delegate the refusal-matrix decision to the deterministic guard (single source of truth for ARCHITECTURE.md §11.5):

bash "$CLAUDE_PLUGIN_ROOT/hooks/scripts/board-mode-guard.sh" worker --discipline <value from step 1>

Inspect the exit code:

  • 0 (ALLOW): continue to Step 4.
  • 2 (NOOP): print the guard's stdout verbatim and stop. The canonical message for this branch is Engineering board: already in worker mode (discipline=<value>). No action taken.
  • 3 (REFUSE): print the guard's stdout verbatim and stop. The canonical messages for this branch are:
    • Engineering board: currently in PM mode. Restart the session to switch to worker mode. No action taken.
    • Engineering board: currently in worker mode (discipline=<existing>). Restart the session to switch to discipline=<requested>. No action taken.
    • Engineering board: currently paused. Run /board-resume first, then /worker-start --discipline <value>. No action taken.

Read the full file on GitHub · 103 lines

Changes

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.

  1. yesterday First seen · 103 lines · 53 tokens per session scan A 37e37c681882

Subscribe to this mod's changes

worker-start is a command published in the GitHub repository GhostlyGawd/engineering-board (0 stars, last pushed yesterday), licensed MIT. It adds 53 tokens to every session and 1,419 once invoked, about $0.0003 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 2026-08-31.