Borrowing it
Nothing to install: this file belongs to katopz/katgpt-rs. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/katopz/katgpt-rs/develop/AGENTS.mdgit clone --depth 1 https://github.com/katopz/katgpt-rsWrote 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/katopz/katgpt-rs/agents-md)<a href="https://agentmods.dev/instructions/katopz/katgpt-rs/agents-md"><img src="https://agentmods.dev/badge/instructions/katopz/katgpt-rs/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.
<a href="https://agentmods.dev/instructions/katopz/katgpt-rs/agents-md"><img src="https://agentmods.dev/badge/instructions/katopz/katgpt-rs/agents-md.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 | $0.27461 | $0.27461 |
| Opus 5.5 | $0.10984 | $0.10984 |
| Sonnet 5.5 | $0.05492 | $0.05492 |
| Haiku 4.5 | $0.02746 | $0.02746 |
Grade A, and why
katgpt-rs AGENTS.md scanned grade A with 1 finding 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 today.
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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
`subprocess.run(..., text=True)` decodes the child's pipe with How it starts
The opening of the file, as written. The whole thing — 1,531 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — katgpt-rs
The global ~/.agents/ rules apply; this file documents repo-local context
that supplements them. History, resolved-issue records, gate narratives,
collision precedents: HISTORY.md (compacted 2026-10-06 — every ## heading
kept; full pre-compaction text in git history). Removed issue files: git
history.
Boundary contract — read BOUNDARY.md first
BOUNDARY.md is the authoritative contract: what this repo
owns, what it does not own (with the correct home for each), the
crate-granular allowlist, and the drift ledger. On any conflict with
prose, BOUNDARY.md wins.
- Domain test: is this a modelless inference primitive with no riir dep (this repo is upstream of everything)? NO → it belongs in another repo; file there.
- Read it before adding any dep, crate, module, System impl, or vocabulary type.
- Enforcement:
../riir-ai/scripts/ci_boundary_contract.sh— undeclared cross-repo dep, drift row without its open issue, contract-vs-measured-graph drift. Run boundary checks VIA theboundary-guardskill, not ad-hoc greps. - Found a violation? File the issue FIRST (
.issues/NNN_boundary_*.md), add the drift row, then fix. Closing the issue removes the row in the same commit.
Modelless-first mandate (the core principle)
This repo ships modelless inference primitives. No training, no backprop, no gradient descent. The only weight mutations allowed at runtime:
- Freeze/thaw — swapping a frozen snapshot (atomic, versioned, BLAKE3-checked).
- Raw/lora hot-swap — a deterministically constructed (not trained)
LoRA overlay via
LoraPair { reader, writer }(Plan 025). - Latent-space updates — direction-vector projections, sigmoid gates, routing tables; latent state, NOT base weights.
MANDATORY: exhaust modelless paths before deferring to riir-train. Before deferring ANY gate, mechanism, or plan task ("this needs training"), check the three paths above (research skill §3.5). Systematic, characterizable biases are modelless-correctable candidates — for a known, named bias, try a deterministic reader-LoRA or freeze-state correction before concluding "needs gradient descent." Canonical failure: AC-Prefix G1 (Plan 313) — the story and the drift history live in HISTORY.md.
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.
- today Changed · +172 tokens per session 6f94dece4429
- yesterday Changed · +32 lines · +649 tokens per session 5f2acade4896
- 2d ago Changed · +10 lines · +179 tokens per session 4b381d99b70c
- 3d ago Changed · +42 lines · +832 tokens per session b144ce4a0d3d
- 4d ago Changed · +46 tokens per session 854b7e3fece2
- 6d ago Changed · +22 lines · +946 tokens per session bd1b31339379
- 8d ago Changed · +12 lines · +277 tokens per session 6aebfacdd21c
- 12d ago Changed · +12 lines · +203 tokens per session 5cf40b1c69cb
- 13d ago Changed · -6 lines · -135 tokens per session bf492777190f
- 14d ago First seen · 1,407 lines · 24,292 tokens per session scan A d2b29a6d582b
katgpt-rs AGENTS.md is an instructions file published in the GitHub repository katopz/katgpt-rs (140 stars, last pushed today), licensed MIT. It adds 27,461 tokens to every session, about $0.1098 per session on Opus 5.5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
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).
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.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.