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 instructions/badrshs/scribe-ai/copilot-instructionsgit clone --depth 1 https://github.com/badrshs/scribe-aiWrote 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/badrshs/scribe-ai/copilot-instructions)<a href="https://agentmods.dev/instructions/badrshs/scribe-ai/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/badrshs/scribe-ai/copilot-instructions.svg" alt="Measured on agentmods" 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 | $0.01680 | $0.01680 |
| Opus 5 | $0.00840 | $0.00840 |
| Sonnet 5 | $0.00336 | $0.00336 |
| Haiku 4.5 | $0.00168 | $0.00168 |
Grade A, and why
scribe-ai copilot-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 4d 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 — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scribe AI - Copilot Instructions
Package identity
- Composer name:
badrshs/scribe-ai| PHP namespace:Badr\ScribeAi - Laravel package (no standalone entrypoint). All classes are registered as singletons in
ScribeAiServiceProvider. - Targets Laravel 11/12 and PHP 8.2+.
Testing Workflow - The "done" Command Loop Veru crocial
Before completing ANY task, you MUST:
- Run
php artisan donecommand and wait for approval - user will test, check your code and potentially modify the command with the next instructions
- If user modifies the command, address their feedback
- Then Run
php .\artisan doneagain - Repeat until the command runs successfully without user modifications
Core architecture
1. Content Pipeline
ContentPipeline sends an immutable ContentPayload DTO through an ordered list of stages using Laravel's Illuminate\Pipeline\Pipeline. Default stage order (configured in config/scribe-ai.php → pipeline.stages):
ScrapeStage → AiRewriteStage → GenerateImageStage → OptimizeImageStage → CreateArticleStage → PublishStage
Writing a stage - implement Contracts\Pipe:
public function handle(ContentPayload $payload, Closure $next): mixed
{
// Transform and continue:
return $next($payload->with(['title' => $newTitle]));
// Or reject (halt pipeline without calling $next):
return $payload->with(['rejected' => true, 'rejectionReason' => 'duplicate']);
}
Mutating the payload - ContentPayload has readonly properties. Always use $payload->with([...]) to produce a new instance; never try to assign properties directly.
One-off custom stage order:
app(ContentPipeline::class)->through([ScrapeStage::class, MyStage::class])->process($payload);
customStages is consumed and cleared after a single process() call.
2. Publisher Manager
PublisherManager is a Strategy + Manager pattern. Built-in drivers: log, facebook, telegram, blogger, wordpress.
Adding a custom driver (in a service provider):
app(PublisherManager::class)->extend('medium', fn(array $config) => new MediumDriver($config));
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.
- 4d ago First seen · 135 lines · 1,680 tokens per session scan A 504ad42ce82c
scribe-ai copilot-instructions.md is an instructions file published in the GitHub repository badrshs/scribe-ai (11 stars, last pushed 3mo ago), licensed MIT. It adds 1,680 tokens to every session, about $0.0084 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 instructions, from other repositories
laravel AGENTS.md
AGENTS.md instructions for laravel/laravel, covering laravel application, prerequisites and agent setup.
laravel CLAUDE.md
Claude Code instructions for laravel/laravel, covering laravel application, prerequisites and agent setup.
livewire-powergrid CLAUDE.md
Claude Code instructions for Power-Components/livewire-powergrid, covering powergrid 7.x ai guidelines, 1. core principles, 2. architecture & theming, 3. views and 4. server-side actions.
LarAgent AGENTS.md
Instructions for MaestroError/LarAgent, covering purpose, key principles, planning, executing plan and testing.
real-estate-laravel copilot-instructions.md
Instructions for liberusoftware/real-estate-laravel, covering lerd, a local php development environment, architecture, dns modes, mcp tools and key conventions.
laravel-mcp-server AGENTS.md
Instructions for opgginc/laravel-mcp-server, covering common commands, testing and quality assurance, mcp tool development, mcp notification development and configuration publishing.