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/thavarshan/filterable/agents-mdgit clone --depth 1 https://github.com/Thavarshan/filterableWhat 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 | $0.00639 | $0.00639 |
| Opus 5 | $0.00319 | $0.00319 |
| Sonnet 5 | $0.00128 | $0.00128 |
| Haiku 4.5 | $0.00064 | $0.00064 |
Grade A, and why
filterable 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 2d 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 — 22 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines
Project Structure & Module Organization
Source code lives under src/Filterable/, centred around the abstract Filter base class and trait-based “concerns” under src/Filterable/Concerns/. Contracts and Eloquent helpers are in src/Filterable/Contracts/ and src/Filterable/Traits/. The Artisan generator and stubs are in src/Filterable/Console/. Executable tooling scripts are stored in bin/, shared media in assets/, and tests (plus fixtures) in tests/, which boot an Orchestra Testbench environment. Composer already maps Filterable\\ and the testing namespaces; place any new factories inside database/factories to maintain autoloading.
Build, Test, and Development Commands
Run composer install to hydrate dependencies, then rely on the Composer scripts: composer lint (delegates to bin/lint.sh for Duster + syntax checks), composer fix (formats via Pint/Duster and saves a log), and composer test (wraps bin/test.sh which accepts flags such as --filter=HandlesRateLimitingTest, --coverage, --parallel, or --test=tests/HandlesFilterablesTest.php). When adding commands, keep the scripts directory executable (chmod +x bin/*.sh).
Coding Style & Naming Conventions
Code follows PSR-12 with four-space indentation enforced by Pint/Duster. Match namespaces to paths (Filterable\\Concerns\\OptimizesQueries, etc.) and stick to the existing naming patterns: suffix traits with the capability (ManagesMemory), concrete filters with Filter, and console commands under Console. Keep feature toggles ($features) and options arrays cohesive—extend the existing map rather than inventing new flags. Prefer expressive method-level docblocks when behaviour is subtle (e.g. cache key generation), otherwise lean on descriptive naming.
Testing Guidelines
The suite is PHPUnit-based (phpunit.xml.dist) and runs inside Orchestra Testbench. Follow the established pattern of placing concern-specific tests at the project root (e.g. CachingTest.php, HandlesRateLimitingTest.php) and keep reusable doubles under tests/Fixtures/. Use partial mocks for collaborators (cache, logger, rate limiter) and prefer data providers or inline anonymous filters for edge cases. Generate coverage with ./bin/test.sh --coverage --filter=Namespace\\Class before shipping complex features, and assert on state via getDebugInfo() when relevant.
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.
- 2d ago First seen · 22 lines · 639 tokens per session scan A 7d9a389c966e
filterable AGENTS.md is an instructions file published in the GitHub repository Thavarshan/filterable (195 stars, last pushed 2mo ago), licensed MIT. It adds 639 tokens to every session, about $0.0032 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
ULogViewer AGENTS.md
Instructions for carina-studio/ULogViewer, covering agents.md, coding style, naming, formatting & structure and fields & properties.
Weavero CLAUDE.md
Claude Code instructions for mjthoraval/Weavero, covering weavero — guidance for ai coding agents, commands, layout, verify, don't guess and hard invariants (each paid for with a real regression).
Weavero AGENTS.md
AGENTS.md instructions for mjthoraval/Weavero: See CLAUDE.md — it is agent-agnostic despite the name (the convention Zotero's own repos use). Workflows live in .claude/skills/, conventions in .claude/rules/, and the design of the whole setup is explained in .claude/README.md.
ULogViewer CLAUDE.md
Instructions for carina-studio/ULogViewer, a project described as: Cross-Platform, Agent-Native Universal Log Viewer.
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.
buildNext
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).