What is tpu_performance_autoresearch_wiki AGENTS.md?

Repository instructions for a wiki about improving TPU model performance through repeated experiments. They describe the shared rules, file locations, compatibility requirements, experiment records, and checks needed before reaching a conclusion.

Use them when editing wiki pages, running or documenting experiments, changing shared agent skills, or updating the repository index and logs.

They keep changes consistent across Claude Code, Codex, and Antigravity and help ensure that experiments are recorded and evaluated in the required way.

How vlasenkoalexey/tpu_performance_autoresearch

Reuse

Borrowing it

Nothing to install: this file belongs to vlasenkoalexey/tpu_performance_autoresearch_wiki. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/vlasenkoalexey/tpu_performance_autoresearch_wiki/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/vlasenkoalexey/tpu_performance_autoresearch_wiki

Made for: Codex, OpenCode.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/vlasenkoalexey/tpu_performance_autoresearch_wiki/agents-md/github.svg)](https://agentmods.dev/instructions/vlasenkoalexey/tpu_performance_autoresearch_wiki/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/vlasenkoalexey/tpu_performance_autoresearch_wiki/agents-md"><img src="https://agentmods.dev/badge/instructions/vlasenkoalexey/tpu_performance_autoresearch_wiki/agents-md/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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/vlasenkoalexey/tpu_performance_autoresearch_wiki/agents-md"><img src="https://agentmods.dev/badge/instructions/vlasenkoalexey/tpu_performance_autoresearch_wiki/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 436 This file is loaded in full into every session.

When invoked 436 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00436	$0.00436
Opus 5	$0.00218	$0.00218
Sonnet 5	$0.00087	$0.00087
Haiku 4.5	$0.00044	$0.00044

Measured 10d ago against content hash c7f6565ae0e2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

tpu_performance_autoresearch_wiki AGENTS.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 10d 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.

AGENTS.md · 43 lines

What it actually says

Codex Instructions

This repository is an autoresearch wiki for TPU model performance optimization. The canonical cross-agent protocol is SCHEMA.md; read it and wiki/index.md before doing wiki, experiment, or analysis work.

Compatibility

Keep .claude/ as the canonical source for shared skills, scripts, and Claude Code agents.
.agents/skills exposes the same skills to Codex. Treat the skill bodies as shared cross-agent workflow definitions.
.codex/ contains Codex-only adapters: MCP config, custom-agent wrappers, and hook wiring. Keep repo behavior additive so Claude and Antigravity remain compatible.

Operating Rules

Follow SCHEMA.md for page formats, wiki paths, lint rules, and the autoresearch loop.
Do not modify raw/ except through the controlled workflows documented in SCHEMA.md and wiki/experiments/program.md.
Update wiki/index.md and the appropriate log after material wiki changes.
Use one experiment page per run.
For executed experiments, XProf and HLO analysis are mandatory before assigning a final verdict.
Prefer repo skills in .agents/skills for lifecycle operations such as hypothesis formulation, experiment start/stop, retrospectives, lint, source ingestion, cluster scans, and AST extraction.

Codex Translation Notes

When a shared skill mentions AskUserQuestion, ask the user directly unless the current Codex surface provides an equivalent structured prompt.
When a shared skill mentions Claude /loop, ScheduleWakeup, Task, TaskStop, or background subagent notifications, translate that step to the available Codex subagent, hook, or automation mechanism. Do not pretend the Claude primitive exists.
When a shared skill references .claude/scripts, use those scripts in place; they are intentionally shared.
When a shared skill references .claude/agents, use the corresponding .codex/agents/*.toml wrapper if running under Codex.

Changes

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.

10d ago First seen · 43 lines · 436 tokens per session scan A c7f6565ae0e2

Subscribe to this mod's changes

tpu_performance_autoresearch_wiki AGENTS.md is an instructions file published in the GitHub repository vlasenkoalexey/tpu_performance_autoresearch_wiki (55 stars, last pushed 6d ago), licensed MIT. It adds 436 tokens to every session, about $0.0022 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.

LLM autoresearch jax knowledge-graph machine-learning mfu

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

tpu_performance_autoresearch_wiki: Instructions file for Codex