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/lst97/claude-code-sub-agents/claude-mdgit clone --depth 1 https://github.com/lst97/claude-code-sub-agentsWrote 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/lst97/claude-code-sub-agents/claude-md)<a href="https://agentmods.dev/instructions/lst97/claude-code-sub-agents/claude-md"><img src="https://agentmods.dev/badge/instructions/lst97/claude-code-sub-agents/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 | $0.02339 | $0.02339 |
| Opus 5 | $0.01170 | $0.01170 |
| Sonnet 5 | $0.00468 | $0.00468 |
| Haiku 4.5 | $0.00234 | $0.00234 |
Grade A, and why
claude-code-sub-agents 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 4d 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 — 266 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Full Stack Development Guidelines
Philosophy
Core Beliefs
- Iterative delivery over massive releases – Ship small, working slices of functionality from database to UI.
- Understand before you code – Explore both front-end and back-end patterns in the existing codebase.
- Pragmatism over ideology – Choose tools and architectures that serve the project’s goals, not personal preference.
- Readable code over clever hacks – Optimize for the next developer reading your code, not for ego.
Simplicity Means
- One clear responsibility per module, class, or API endpoint.
- Avoid premature frameworks, libraries, or abstractions.
- While latest and new technology is considerable, stable and efficient should be prioritized.
- If your integration flow diagram needs an explanation longer than 3 sentences, it’s too complex.
Process
1. Planning & Staging
Break work into 3–5 cross-stack stages (front-end, back-end, database, integration). Document in IMPLEMENTATION_PLAN.md:
## Stage N: [Name]
**Goal**: [Specific deliverable across the stack]
**Success Criteria**: [User story + passing tests]
**Tests**: [Unit, integration, E2E coverage]
**Status**: [Not Started|In Progress|Complete]
- Update status after each merge.
- Delete the plan file after all stages are verified in staging and production.
2. Implementation Flow
- Understand – Identify existing patterns for UI, API, DB, and CI/CD.
- Test First – For back-end, write API integration tests; for front-end, write component/unit tests.
- Implement Minimal – Just enough code to pass all tests.
- Refactor Safely – Clean code with test coverage at 60%+ for changed areas.
- Commit Clearly – Reference plan stage, include scope (front-end, back-end, DB).
3. When Stuck (Max 3 Attempts)
- Document Failures – Include console logs, stack traces, API responses, and network traces.
- Research Alternatives – Compare similar solutions across different tech stacks.
- Check Architecture Fit – Could this be a UI-only change? A DB query rewrite? An API contract change?
- Try a Different Layer – Sometimes a front-end bug is a back-end response problem.
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.
- 4d ago First seen · 266 lines · 2,339 tokens per session scan A d03da932184d
claude-code-sub-agents CLAUDE.md is an instructions file published in the GitHub repository lst97/claude-code-sub-agents (1,674 stars, last pushed 1y ago), licensed MIT. It adds 2,339 tokens to every session, about $0.0117 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 instructions, from other repositories
my-claude-code-setup AGENTS.md
Instructions for centminmod/my-claude-code-setup, covering always start with these commands for common tasks, or, directory/file exploration, first choice - list all files/dirs recursively and for current directory only.
my-claude-code-setup CLAUDE.md
Instructions for centminmod/my-claude-code-setup, covering claude.md, project overview, claude.md templates, tools and rules dependency.
my-claude-code-setup GEMINI.md
Instructions for centminmod/my-claude-code-setup, covering gemini.md, ai guidance, memory bank system, core context files and project overview.
pi-subagents AGENTS.md
AGENTS.md instructions for tintinweb/pi-subagents, covering development rules, conversational style, code quality, documentation and commands.
taskflow AGENTS.md
AGENTS.md instructions for heggria/taskflow, covering agents.md, project overview, architecture, key concepts and phase types (12 total).
claude-code-riper-5 AGENTS.md
Instructions for tony/claude-code-riper-5, covering agents.md — claude-code-riper-5, git commit standards, documentation standards, code blocks and show the last ten commits as a graph.