A build command that carries out an approved software plan in waves, where each story is a small piece of work assigned to a worker. A wave is a group of independent stories that can run at the same time.

Use it to dispatch approved stories in order, run parallel work safely, and create one commit for each completed story.

It checks the plan for conflicting files, missing paths, and weak verification before work starts, reducing accidental overwrites and misleading successful results.

How Black-coffe/vulyk configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to Black-coffe/vulyk. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/Black-coffe/vulyk/main/.claude/commands/vulyk-build.md

Clone the repo

git clone --depth 1 https://github.com/Black-coffe/vulyk

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/commands/black-coffe/vulyk/vulyk-build/github.svg)](https://agentmods.dev/commands/black-coffe/vulyk/vulyk-build)

Your own site

<a href="https://agentmods.dev/commands/black-coffe/vulyk/vulyk-build"><img src="https://agentmods.dev/badge/commands/black-coffe/vulyk/vulyk-build/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/commands/black-coffe/vulyk/vulyk-build"><img src="https://agentmods.dev/badge/commands/black-coffe/vulyk/vulyk-build.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 42 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 2,996 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00042	$0.02996
Opus 5.5	$0.00017	$0.01198
Sonnet 5.5	$0.00008	$0.00599
Haiku 4.5	$0.00004	$0.00300

Measured today against content hash 589bfc7a3cc7, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

vulyk-build 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 today.

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.

.claude/commands/vulyk-build.md · 135 lines

How it starts

The opening of the file, as written. The whole thing — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Build: "$ARGUMENTS" (default: the newest spec under docs/specs/ whose plan.md carries **Briefed:** or **Approved:** and no **Branch:**).

plan.md's **Tier:** picks the path: Tier 1-2 solo, Tier 3-4 hive. Every cycle.sh verb prints one JSON object as its last stdout line; decide from its ok, next and status, never from prose. Bash calls that run advance or close-story carry timeout: 600000. After each call print one line: the verb, advance's steps, and next.

Solo (Tier 1-2)

No driver, no clerk, no court. Loop:

bash scripts/cycle.sh advance docs/specs/<slug>. It runs every mechanical step (branch, open-round, judge, the repair story) up to the next thing only an agent can do, and names it next.
build:<W>: take status.wave_stories in order. For a red: required story, first write its test (the file ## Files names) and red_reason:, then run bash scripts/cycle.sh stamp-red <file> (no --stamp), and only then implement: no implementation edit before the receipt, never a staged break to produce a red. Its exit-4 messages are the workers' (.claude/agents/worker-code.md step 6) and handled the same way; where a worker returns NEEDS_CONTEXT, stop for the owner. For each story, implement it yourself (its ## Files only; a bug fix runs bash scripts/symbol-map.sh <file> --story <story> on a root file before editing it, greps only what it cannot vouch for (names marked textual, every caller of a file it reports imports not parsed) and fixes the shared code once; targeted checks while working), add ## Implementation notes, set returned: DONE, and run bash scripts/cycle.sh close-story <file> --commit. On exit 4, read the error, fix, rerun; after three failed reruns write ## Findings and stop for the owner. A repair story (# Repair round <n>) is built the same way; its ## Findings are the conditions to meet. Back to 1.
dispatch:review: dispatch lead-review with the Agent tool, with model: <r> from r="$(bash scripts/model-profile.sh --role review)" only when the profile (bash scripts/model-profile.sh) is not sonnet-builds, else no model parameter: Review <slug>, round <round>. Spec: <spec>. Branch <branch> at <head>. Report: .vulyk/reports/<slug>/round-<round>/review.attempt-<k>.md with k = status.seat_attempt.review; from round 2 add Since: <since>. Previous round: <spec>/council/round-<round-1>. Then bash scripts/cycle.sh advance docs/specs/<slug> --ingest. If next is still dispatch:review, the report was rejected: dispatch once more with the rejected entry's error appended, and --ingest again. Then act on the new next. A dispatch that returns the Fable-limit refusal (You've reached your Fable limit / model_requires_usage_credits) gets the Fable-limit fallback (below), <where> = round-<round>/review, before --ingest.
Any other next is terminal (below).

Read the full file on GitHub · 135 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.

today Changed 589bfc7a3cc7
yesterday Changed · +5 lines 7bd363562a01
2d ago Changed · +26 lines 08d265d8c668
3d ago Changed · +1 lines 1f9e9abd6d07
10d ago Changed · +9 lines 45efedee36b3
13d ago Changed · +1 lines b1e8f4f5a54e
21d ago Changed · +3 lines 384e21d577ed
1mo ago Changed · +4 lines 1d7ffcc4c7b5
1mo ago Changed · +2 lines b15ea6692d47
1mo ago First seen · 84 lines · 42 tokens per session scan A cedb6e4e8935

Subscribe to this mod's changes

vulyk-build is a command published in the GitHub repository Black-coffe/vulyk (2 stars, last pushed today), licensed MIT. It adds 42 tokens to every session and 2,996 once invoked, about $0.0002 per session on Opus 5.5. 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.

Claude Code Planning agentic-coding ai-agents ai-framework anthropic claude

vulyk: Command for Claude Code

Borrowing it

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Solo (Tier 1-2)

What this file has done since we first saw it

Other commands, from other repositories

audit-agents-skills

security-audit

audit-quiz-coverage

worktree

sync

diff-docs