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 instructions/lantisprime/claude-sdlc/claude-mdgit clone --depth 1 https://github.com/lantisprime/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/instructions/lantisprime/claude-sdlc/claude-md)<a href="https://agentmods.dev/instructions/lantisprime/claude-sdlc/claude-md"><img src="https://agentmods.dev/badge/instructions/lantisprime/claude-sdlc/claude-md.svg" alt="Measured on agentmods" 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 | $0.02840 | $0.02840 |
| Opus 5 | $0.01420 | $0.01420 |
| Sonnet 5 | $0.00568 | $0.00568 |
| Haiku 4.5 | $0.00284 | $0.00284 |
Grade A, and why
claude-sdlc CLAUDE.md 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 5d 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 — 190 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Context for Claude Code sessions working on this repository — the SDLC plugin itself. This file is read on startup. Keep it short; keep it honest.
What this repo is
A Claude Code plugin that enforces an 8-phase SDLC workflow with human-in-the-lead discipline. It is not application code — it is instructions, hooks, templates, and agents that shape how other Claude Code sessions work in other repos.
Read README.md for the quick-start and docs/SDLC.md for the full phase reference before making substantive changes.
Design intent — read this before "improving" anything
These principles are essential. Changes that violate them usually feel like simplifications but aren't:
-
Human in the lead, always. Every phase ends at a signed gate file. Subagents, hooks, and MCP connectors propose — humans approve. Do not add an "auto-approve" convenience. Do not let a subagent advance a phase. Do not let observation of past behavior substitute for a fresh human confirmation.
-
Reduce cognitive load, always. The human is the bottleneck this plugin is designed to protect. Additions to artifacts, prompts, hook messages, and subagent output must make the human's decision easier — less to read, less to reconcile, clearer next step. If a change adds surface area without removing more elsewhere, it's the wrong change.
-
Plan before code.
plan-gate.shblocks Edit/Write when no plan exists. This is the single most important rule in the plugin. If it seems to be firing "too often," the fix is better planning, not a looser hook. -
Surgical edits. Skills and hooks together enforce that only plan-listed files and functions are modified. Adjacent functions are never touched. "While I'm here" cleanups are a known failure mode, not a virtue.
-
Work-item traceability. Every build references a REQ ID, an issue ticket, or a signed CR. Don't add a "quick fix" escape hatch that bypasses this.
-
Graceful degradation. Missing Git? Missing ticket system? Missing observability platform? The plugin writes markdown/JSON artifacts locally and surfaces the gap. It never silently skips a check.
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.
- 5d ago First seen · 190 lines · 2,840 tokens per session scan A 490f0c16ea5d
claude-sdlc CLAUDE.md is an instructions file published in the GitHub repository lantisprime/claude-sdlc (3 stars, last pushed 4mo ago), licensed MIT. It adds 2,840 tokens to every session, about $0.0142 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.
Other instructions, from other repositories
waterfall AGENTS.md
Instructions for mgallet92i/waterfall, covering agents.md — développer le framework waterfall, philosophie, carte du repo, workflow — la state machine, phase par phase and vue d'ensemble.
waterfall CLAUDE.md
Instructions for mgallet92i/waterfall, a project described as: Multi-agent specification-driven development (SDD) framework for Claude Code.
lead-dev-os CLAUDE.md
Instructions for CaptainMe-AI/lead-dev-os, covering claude.md, project, terminology, plugin architecture and repository structure.
embedded-systems-architect AGENTS.md
Instructions for Zachariah9420/embedded-systems-architect: This repository ships an agent skill for embedded-systems and edge-AI development work.
qa-orchestra CLAUDE.md
Claude Code instructions for Anasss/qa-orchestra, covering qa orchestra — claude code instructions, what this project is, quick start, 1. clone into your workspace and 2. fill in your project context.
caddis-plugin browser-first.instructions.md
Direct integrated-browser actions must use native browser tools first.