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/graykode/abtop/agents-mdgit clone --depth 1 https://github.com/graykode/abtopWhat 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 | $0.05449 | $0.05449 |
| Opus 5 | $0.02724 | $0.02724 |
| Sonnet 5 | $0.01090 | $0.01090 |
| Haiku 4.5 | $0.00545 | $0.00545 |
Grade B, and why
abtop AGENTS.md scanned grade B with 1 finding 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
`abtop --setup` automates this: creates a script at `~/.claude/abtop-statusline.sh` that writes rate limit JSON to `~/.claude/abtop-rate-limits.json`, and registers it in `~/.claude/settings.json`. How it starts
The opening of the file, as written. The whole thing — 410 lines — stays where its author put it; the contents beside it link to each section on GitHub.
abtop
AI agent monitor for your terminal. Like btop++, but for AI coding agents.
Supports Claude Code, Codex CLI, and OpenCode sessions.
Language Policy
English is mandatory for all project-facing work and communication.
- Write all source code, comments, tests, fixtures, documentation, examples, configuration text, scripts, and user-facing strings in English.
- Use English for every GitHub artifact: issue titles and bodies, issue comments, pull request titles and descriptions, review comments, commit messages, branch names, release notes, changelogs, discussions, labels, milestones, and workflow or CI messages.
- Do not use non-English text in repository content or GitHub communication unless it is an exact external identifier, a required protocol value, or a direct quote needed for context.
- When quoting or preserving non-English input, add an English explanation and keep the non-English text as short as possible.
- If a contributor opens an issue, comment, or review in another language, respond in English and continue the thread in English.
Architecture
src/
├── main.rs # Entry, terminal setup, event loop, --setup flag
├── app.rs # App state, tick logic, key handling, summary generation
├── setup.rs # StatusLine hook installation (abtop --setup)
├── ui/
│ └── mod.rs # All panels in single file: header, context, quota,
│ # tokens, projects, ports, sessions, footer
├── collector/
│ ├── mod.rs # MultiCollector orchestration, orphan port detection
│ ├── claude.rs # Claude Code: session discovery, transcript parsing
│ ├── codex.rs # Codex CLI: session discovery via ps+lsof, JSONL parsing
│ ├── opencode.rs # OpenCode: session discovery via ps + SQLite DB parsing
│ ├── process.rs # Child process tree (ps) + open ports (lsof) + git stats
│ └── rate_limit.rs # Rate limit file reading (~/.claude/abtop-rate-limits.json)
└── model/
├── mod.rs # Re-exports
└── session.rs # AgentSession, SessionStatus, RateLimitInfo,
# ChildProcess, OrphanPort, SubAgent
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 · 410 lines · 5,449 tokens per session scan B 045975794e89
abtop AGENTS.md is an instructions file published in the GitHub repository graykode/abtop (3,480 stars, last pushed 1mo ago), licensed MIT. It adds 5,449 tokens to every session, about $0.0272 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
dev-agent-skills CLAUDE.md
Instructions for fvadicamo/dev-agent-skills, covering claude.md, project overview, architecture, key file: marketplace.json and skill anatomy.
Ocul-PM CLAUDE.md
Instructions for bunhine0452/Ocul-PM, covering claude.md, what this is, commands, single frontend test and rust (run from src-tauri/).
vix AGENTS.md
AGENTS.md instructions for get-vix/vix, covering agents.md, project overview, architecture, instance control channel and development commands.
ob-1 AGENTS.md
Instructions for Overbrilliant/ob-1, covering agents.md, project index, stack, commands and key files.
RemoClaw CLAUDE.md
Instructions for quangtam/RemoClaw, covering remoclaw — agent context, what is this?, architecture, file structure and how to run.
lerim AGENTS.md
AGENTS.md instructions for nablo-io/lerim, covering lerim, summary, current architecture, trace format and parsing and how to query existing context.