Borrowing it
Nothing to install: this file belongs to rjmurillo/ai-agents. 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/rjmurillo/ai-agents/main/.github/instructions/claude-model-patches.instructions.mdgit clone --depth 1 https://github.com/rjmurillo/ai-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/rjmurillo/ai-agents/claude-model-patches)<a href="https://agentmods.dev/instructions/rjmurillo/ai-agents/claude-model-patches"><img src="https://agentmods.dev/badge/instructions/rjmurillo/ai-agents/claude-model-patches/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/rjmurillo/ai-agents/claude-model-patches"><img src="https://agentmods.dev/badge/instructions/rjmurillo/ai-agents/claude-model-patches.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 | $0.01322 | $0.01322 |
| Opus 5 | $0.00661 | $0.00661 |
| Sonnet 5 | $0.00264 | $0.00264 |
| Haiku 4.5 | $0.00132 | $0.00132 |
Grade A, and why
ai-agents claude-model-patches.instructions.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 5d 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 — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude Model Behavioral Patches
These nudges are tuned for the Claude model family (Anthropic models). They apply whenever Claude is the runtime model, regardless of harness (Claude Code CLI, GitHub Copilot CLI invoking Anthropic, the Anthropic API, an IDE extension wrapping Claude). The patches encode behaviors that the model tends to drift from without explicit prompting.
Loading vs. obeying. In the canonical source under .claude/rules/ this file's frontmatter sets priority: critical, meaning the rule loader MUST include the file in every session's context. That key is intentionally stripped from the generated Copilot instruction mirrors (.github/instructions/*.instructions.md, src/copilot-cli/instructions/*.instructions.md) because those trees use a different frontmatter contract (applyTo only). The CONTENT priority of each patch against other rules is set by the Precedence section below. Critical-to-load is not the same as critical-to-obey; the next section sets the obey order.
Precedence
These patches are subordinate to:
- Skill workflows (any
.claude/skills/<name>/SKILL.mdinstruction in this repo, or.github/skills/<name>/SKILL.mdor any installed plugins). - STOP points and confirmation gates inside skills.
AskUserQuestion(or harness equivalent) blocking on user decision.- Plan-mode safety in Claude Code.
- The
/shipreview gate and equivalent push-time checks.
When a patch below conflicts with one of those, the skill or gate wins. Treat the patches as preferences in those conflicts. Outside of those conflicts the patches still apply: they encode default behavior that the model otherwise drifts from.
Todo-list Discipline
When working through a multi-step plan, mark each task complete individually as you finish it. Do not batch-complete at the end. If a task turns out to be unnecessary, mark it skipped with a one-line reason.
Reason: a batched complete-everything-at-the-end pattern hides progress from the user and from any orchestrator watching the run. If the model crashes or the session ends mid-run, the todo list still reflects reality up to the last completed step. Batched updates make every recovery start from zero.
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.
- 5d ago First seen · 88 lines · 1,322 tokens per session scan A 8bfb83d15eb5
ai-agents claude-model-patches.instructions.md is an instructions file published in the GitHub repository rjmurillo/ai-agents (45 stars, last pushed yesterday), licensed MIT. It adds 1,322 tokens to every session, about $0.0066 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 2026-09-03.
Other instructions, from other repositories
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.
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.
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).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
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.
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.