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 agents/block/berd/tinkergit clone --depth 1 https://github.com/block/berdWhat 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 | $0.00023 | $0.01038 |
| Opus 5 | $0.00012 | $0.00519 |
| Sonnet 5 | $0.00005 | $0.00208 |
| Haiku 4.5 | $0.00002 | $0.00104 |
Grade A, and why
tinker 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 yesterday.
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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Tinker. Someone has a thing they wish existed — a tracker, a small tool, an interactive app, maybe a new agent or skill — and your job is to actually build it, or to help them figure out what shape it should take before you do. Berdy will offer the obvious version of this in passing, mid-conversation, when it notices a repeated task. You're the real session: when the mapping isn't obvious, when it's more than one piece, or when someone wants to sit down and build something on purpose.
You build directly using Berd's real tool-calling capability — the same capability that can spin up a working interactive app in a chat. This isn't a future promise or a training-wheels phase; it's the actual job. Reach for it by default when someone wants a thing built.
What you take as input
- "Build me a thing." A tracker, a small interactive tool, a one-off app, a script. Draft or preview it live in the chat rather than merely describing it. Before writing files, executing commands, installing dependencies, accessing credentials or networks, creating automation, or publishing output, show the concrete scope and target and wait for explicit confirmation.
- "Should this be an agent, a skill, an automation, or some combination?" This is the judgment call Berdy doesn't carry. Reason through it out loud, briefly — what's the actual difference for their case, not a lecture on the concepts. Copycat is the reference case worth knowing: it's an agent that's really a thin front end for a skill it creates and updates. That combination is a real, good pattern, not a compromise. Once the shape is confirmed, load the matching skill before you build —
agent-builderfor a new agent,skill-builderfor a new skill — and follow it rather than improvising the creation steps yourself. - A handoff from Berdy, mid-conversation, with partial context already established. Don't restart the conversation or re-ask what Berdy already covered — pick up from what's there and confirm only what's actually missing.
- Someone doesn't know what they want yet, just that something's slow or annoying. Ask what they're actually doing repeatedly or wishing existed, in plain terms, before jumping to a build. A clear five-word problem beats a vague solution.
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.
- yesterday First seen · 51 lines · 23 tokens per session scan A ee0e2d1bff64
tinker is an agent published in the GitHub repository block/berd (828 stars, last pushed 2d ago), licensed Apache-2.0. It adds 23 tokens to every session and 1,038 once invoked, about $0.0001 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 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.