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/rem42/scraper/agents-mdgit clone --depth 1 https://github.com/rem42/scraperWhat 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 | $0.01364 | $0.01364 |
| Opus 5 | $0.00682 | $0.00682 |
| Sonnet 5 | $0.00273 | $0.00273 |
| Haiku 4.5 | $0.00136 | $0.00136 |
Grade A, and why
scraper 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 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 — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
But et vue d'ensemble
Ce package fournit un petit framework de "scraper" réutilisable :
- Définition d'une requête (classe Request) annotée par l'attribut
#[Scraper(...)]. - Un
Clientqui transforme la Request en appel HTTP (via unHttpClientInterfaceinjecté), puis instancie la classeApicorrespondante pour parser la réponse.
Points d'entrée importants
src/Client.php— méthode principale :Client::send(ScraperRequest $request).- construit les options HTTP à partir des interfaces implémentées par la Request (headers, query, body, json, auth). Voir
buildOptions(). - récupère l'attribut
ScraperviaExtractAttribute::extract()et appelle le endpoint. - instancie la classe Api correspondante en remplaçant "Request" par "Api" dans le nom de classe (voir
getApiReflectionClass()).
- construit les options HTTP à partir des interfaces implémentées par la Request (headers, query, body, json, auth). Voir
Patterns et conventions spécifiques
- PSR-4 autoload: namespace racine
Scraper\Scraper\->src/(voircomposer.jsonautoload). - Naming convention Request->Api : si vous créez
FooRequest, fournissezFooApi(extendsApi\AbstractApi) — leClients'en sert automatiquement. - Attributs PHP: les Requests sont annotées par
#[Scraper(...)](voirsrc/Attribute/Scraper.php) : contient method, scheme, host, path. - Variable substitution dans
path:{name}est remplacé par l'appelgetName()sur l'objet Request (voirsrc/Attribute/ExtractAttribute.php). - Options HTTP: implémentez les interfaces de
src/Request/pour activer les options automatiques :RequestHeaders->getHeaders()→ optionheadersRequestQuery->getQuery()→ optionqueryRequestBody/RequestBodyJson->getBody()/getJson()→ optionbody/jsonRequestAuthBearer/RequestAuthBasic->getBearer()/getAuthBasic()→auth_bearer/auth_basic
Composants clés (exemples)
src/Client.php— orchestrateur principal, error handling et mapping Request->Api.src/Attribute/ExtractAttribute.php— lit les attributs#[Scraper], fusionne héritages et remplace les variables danspath.src/Factory/SerializerFactory.php— création duSerializerSymfony utilisé par les API pour dénormaliser les réponses.src/Request/ScraperRequest.php— classe de base pour les Requests (ssl, auth basic helpers).
Dépendances et intégrations
- Dépendances runtime (extraites de
composer.json):symfony/http-client-contracts(injectionHttpClientInterface)symfony/serializer-pack(Serializer + normalizers)
- Extensions suggérées:
ext-json,ext-simplexml,ext-soap(voirsuggestdanscomposer.json).
Flux d'exécution succinct
- L'utilisateur instancie une Request (ex:
new FooRequest()), éventuellement configure headers/query/body. Client::send($request)→ExtractAttribute::extract($request)pour construire URL/méthode.HttpClientInterface->request(...)est appelé avec les options produites parbuildOptions().- Une classe
FooApi(remplacement Request->Api) est instanciée et sa méthodeexecute()est retournée.
Workflows dev (build / test / QA)
- Tests unitaires: depuis la racine du package ou du monorepo, lancer:
composer run unit-test
# ou
./vendor/bin/phpunit -c packages/scraper/phpunit.xml
- Analyse statique:
composer run static-analysis(phpstan) — configuration :./vendor/bin/phpstan analyse src --level=max. - Format / style:
composer run code-style-checketcomposer run code-style-fix(php-cs-fixer).
Conseils pratiques pour les agents/automates
- Pour ajouter un nouveau scraper pour un site :
- Créer
src/Request/MySiteRequest.phpextendsScraperRequestet ajoutez les getters utilisés par{...}dans le path. - Annoter la classe par
#[Scraper(method: Method::GET, scheme: Scheme::HTTPS, host: 'example.com', path: '/foo/{id}')]. - Créer
src/Api/MySiteApi.phpextendsApi\AbstractApiet implémenterexecute(): object|array|bool|stringpour parserResponseInterfacevia$this->serializer.
- Créer
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 · 94 lines · 1,364 tokens per session scan A 086e8ec0ab1b
scraper AGENTS.md is an instructions file published in the GitHub repository rem42/scraper (5 stars, last pushed 22d ago), licensed MIT. It adds 1,364 tokens to every session, about $0.0068 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 2026-08-31.
Other instructions, from other repositories
WebReaper CLAUDE.md
Instructions for alex-on-ai/WebReaper, covering claude.md, what this is, commands, test reality and toolchain.
frankenwails CLAUDE.md
Instructions for johanjanssens/frankenwails, covering frankenwails, architecture, build, key patterns and wails assetserver (no http server).
docshark AGENTS.md
AI Agent Guidelines & Coding Standards for DocShark.
Browser4 CLAUDE.md
Instructions for platonai/Browser4, covering browser4 — project context for claude, architecture, key dispatch chain (cli → browser), batch commands and e2e test structure.
FreshRSS _general.instructions.md
Instructions for FreshRSS/FreshRSS, covering freshrss general instructions, kiss principle: keep it simple, style and formatting and spelling.
spring-ai-agentcore AGENTS.md
Instructions for spring-ai-community/spring-ai-agentcore, covering agents.md, project overview, architecture, key components and artifact store classes.