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.
git clone --depth 1 https://github.com/AratKruglik/claude-sdlcWrote 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.
[](https://agentmods.dev/commands/aratkruglik/claude-sdlc/start)<a href="https://agentmods.dev/commands/aratkruglik/claude-sdlc/start"><img src="https://agentmods.dev/badge/commands/aratkruglik/claude-sdlc/start/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.
<a href="https://agentmods.dev/commands/aratkruglik/claude-sdlc/start"><img src="https://agentmods.dev/badge/commands/aratkruglik/claude-sdlc/start.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00034 | $0.01995 |
| Opus 5.5 | $0.00014 | $0.00798 |
| Sonnet 5.5 | $0.00007 | $0.00399 |
| Haiku 4.5 | $0.00003 | $0.00199 |
Grade A, and why
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 11d 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 — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/sdlc:start
Single entry point for the SDLC pipeline.
Execution model
This pipeline runs synchronously in the current Claude Code session — it is not a detached or autonomous background process. You stay engaged through:
- Steps 0-2 (dependency preflight, stack detection, git-flow detection, skip-rule analysis): a series of read-only
Glob/Read/gitcalls, each of which may prompt for tool-use permission depending on your settings. - The branch gate (Step 0b-git): the pipeline reports the detected branching model, the classified task type and a proposed branch name, then asks you to create / continue / rename / change type / abort before anything is written.
- Phase boundaries: each phase prints an announcement banner and the orchestrator waits for its result before continuing.
- The development phase's plan/approval gate: the architect writes an implementation plan, then the orchestrator stops and asks you to approve / request changes / abort before any code is written.
Setting SDLC_NONINTERACTIVE=true (see Headless mode) removes interactive prompts, but the pipeline still executes in-session — it does not become a background job.
The final documentation phase autonomously opens a Pull Request via gh pr create (or the GitHub MCP equivalent) once all prior phases complete.
A run survives a lost session: the orchestrator checkpoints its state at every phase boundary in .claude/.sdlc-run-active.json, and /sdlc:start --resume continues from the first unfinished phase without re-running (or re-paying for) the completed ones. See pipeline-orchestrator/SKILL.md Step R.
Hook fast path
Typing /sdlc:start fires the plugin's UserPromptSubmit hook, which runs scripts/run-plan.sh and injects a precomputed run plan (dependency preflight, stack, git-flow detection and branch proposal, diff signals, skip rules, workflow resolution, task slug) before the orchestrator starts. The orchestrator then prints the 🔌 / 🎯 / 🌿 / ✂️ / workflow: blocks from that plan instead of re-deriving them with a dozen Glob/Read/git calls on the session model. The branch gate and the plan approval gate are still asked interactively. Without jq, or with hooks disabled, the orchestrator falls back to its inline algorithm — same output, more tool calls.
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.
- 11d ago Changed · +4 lines 2928f9f51721
- 1mo ago Changed · +5 lines e25c5433b662
- 1mo ago Changed · +1 lines 8ebc626d5996
- 1mo ago First seen · 107 lines · 34 tokens per session scan A 1d5c0f17ec03
start is a command published in the GitHub repository AratKruglik/claude-sdlc (53 stars, last pushed 2d ago), licensed MIT. It adds 34 tokens to every session and 1,995 once invoked, about $0.0001 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 2026-09-03.
Other commands, from other repositories
template
Manage issue templates for streamlined issue creation.
fest-show
Show festival progression (in-progress tasks, roadmap, and dependency view).
add-note
Add an internal or external note to a ConnectWise PSA ticket.
dispatcher
Use when you want the orchestrator to pick the next repo to work on across your whole portfolio — it enumerates candidate repos below the confinement root, resolves free/busy from each repo's session.lock lease, ranks the FREE ones by backlog priority × staleness × readiness, recommends the single most worthwhile one…
sync-linear
Sync current work with Linear ticket status.
workpm
A project-management workflow for coordinating multiple AI workers through five stages. It includes task assignment, shared activity logs, worker replacement, and final checks.