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/itsmemeworks/adhx/agents-mdgit clone --depth 1 https://github.com/itsmemeworks/adhxWhat 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 | $0.00978 | $0.00978 |
| Opus 5 | $0.00489 | $0.00489 |
| Sonnet 5 | $0.00196 | $0.00196 |
| Haiku 4.5 | $0.00098 | $0.00098 |
Grade A, and why
adhx 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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Orientation for AI agents (and humans) starting work in this repo with no prior context.
What this project is
ADHX (adhx.com) — "Save it. Lose it. Find it." An open-source bookmark manager for people who save everything and read nothing. It syncs X/Twitter bookmarks, previews and saves Instagram Reels, TikToks, and YouTube Shorts via a URL-prefix trick (x.com/... → adhx.com/...), and turns the community's anonymous save/preview activity into public, crawlable pages (/trending). Next.js 16 App Router, SQLite + Drizzle, deployed on Fly.io (staging adhx.fly.dev, production adhx.com).
Strategically, ADHX's growth engine is SEO/GEO: x.com is effectively uncrawlable, so ADHX preview pages act as the indexable, structured mirror of social content — every save creates a public long-tail landing page. Features that add per-user variable cost (metered AI) are deliberately out of scope for the free product.
How to get context (read in this order)
README.md— product tour, features, quick start.ARCHITECTURE.md— narrative walkthrough: data flow, auth, preview model, database design.CLAUDE.md— the deep reference: conventions, security invariants (SSRF allowlists, token refresh coalescing, multi-user isolation, activity anonymity), API routes, gotchas. Binding for all agents, not just Claude Code.docs/WORKLOG.md— append-only, newest-first log of what's been done recently and why, including in-flight work and open follow-ups. Read at least the most recent entries before starting; this is where you learn what changed after the docs above were written.
Context protocol (keep the loop alive)
After completing any substantive piece of work (feature, fix with a lesson, architectural decision, reverted experiment), append an entry to docs/WORKLOG.md — newest first, dated, ≤10 lines: what was done, why, current state (shipped/in-flight/reverted), and follow-ups. Never rewrite or delete old entries. This is how the next agent — on a fresh branch, with no conversation history — inherits your context.
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 · 55 lines · 978 tokens per session scan A be96748afc98
adhx AGENTS.md is an instructions file published in the GitHub repository itsmemeworks/adhx (15 stars, last pushed 4d ago), licensed MIT. It adds 978 tokens to every session, about $0.0049 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
karakeep AGENTS.md
AGENTS.md instructions for karakeep-app/karakeep, covering karakeep project overview, project overview, project structure, applications (apps/) and packages (packages/).
karakeep CLAUDE.md
Claude Code instructions for karakeep-app/karakeep, a project described as: A self-hostable bookmark-everything app (links, notes and images) with AI-based automatic tagging and full text search.
karakeep GEMINI.md
Gemini CLI instructions for karakeep-app/karakeep, a project described as: A self-hostable bookmark-everything app (links, notes and images) with AI-based automatic tagging and full text search.
Siftly CLAUDE.md
Instructions for viperrcrypto/Siftly, covering siftly, quick setup, ai authentication — no api key needed, key commands and project structure.
core CLAUDE.md
Claude Code instructions for module-federation/core, covering claude task parallelization & development guidelines, core principles, maximum parallelization strategy, universal parallel task patterns and for any codebase task.
Codex-Skills AGENTS.md
Instructions for TheGoat395/Codex-Skills, covering repository operating notes, required checks, evidence rules and change hygiene.