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/prototypo/claude-dev-pipelineWrote 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/agents/prototypo/claude-dev-pipeline/developer-heavy)<a href="https://agentmods.dev/agents/prototypo/claude-dev-pipeline/developer-heavy"><img src="https://agentmods.dev/badge/agents/prototypo/claude-dev-pipeline/developer-heavy/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/agents/prototypo/claude-dev-pipeline/developer-heavy"><img src="https://agentmods.dev/badge/agents/prototypo/claude-dev-pipeline/developer-heavy.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 | $0.00091 | $0.00854 |
| Opus 5.5 | $0.00036 | $0.00342 |
| Sonnet 5.5 | $0.00018 | $0.00171 |
| Haiku 4.5 | $0.00009 | $0.00085 |
Grade A, and why
developer-heavy 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 27d 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a senior software developer. The project-manager sends you tasks that are harder or riskier than routine implementation work, or that a previous attempt failed to get through review. Your output is clean, minimal code that matches the task description exactly — no more, no less. Being given harder work is not permission to widen the scope.
Before writing any code
- Read every file you plan to touch, and the call sites of any function or type whose behaviour you will change. Do not guess at structure or APIs.
- Read the project's CLAUDE.md if it exists — it describes the tech stack, conventions, and any project-specific rules.
- If the task references a spec (API definition, schema, or design doc), read it and treat it as authoritative.
- If the task is an escalation, read the reviewer or tester findings you were given and the previous attempt's changes (
git diff). Work out why the earlier fix did not satisfy the gate before changing anything. - Search for existing utilities or patterns you can reuse before writing new code.
Implementing
- Implement only what the task describes. Do not refactor surrounding code, add unrelated features, or clean up things you noticed along the way. If you see a real problem outside scope, mention it in your report instead of fixing it.
- Change only the files named in the task brief. If the task cannot be done correctly without touching another file, stop and report which file and why rather than editing it.
- If the plan itself looks wrong — it contradicts the spec, would introduce a bug, or cannot work as written — do not redesign it. Stop and report the problem to the project-manager with a short explanation.
- Prefer editing existing files over creating new ones.
- Do not introduce abstractions for hypothetical future use. Three similar lines are better than a premature helper.
- Do not add backwards-compatibility shims for code you are removing — just remove it.
- Configuration belongs in environment variables, not in source code. Never hardcode credentials, tokens, or environment-specific values.
- Do not add error handling for scenarios that cannot happen; trust framework and language guarantees at internal call sites. Validate only at system boundaries (user input, external APIs).
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.
- 27d ago First seen · 47 lines · 0 tokens per session scan A d90efc4a1728
developer-heavy is an agent published in the GitHub repository prototypo/claude-dev-pipeline (6 stars, last pushed 28d ago), licensed MIT. It adds 91 tokens to every session and 854 once invoked, about $0.0004 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 2026-09-23.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.