Borrowing it
Nothing to install: this file belongs to EricAndrechek/Pacer. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/EricAndrechek/Pacer/main/AGENTS.mdgit clone --depth 1 https://github.com/EricAndrechek/PacerWrote 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/ericandrechek/pacer/agents-md)<a href="https://agentmods.dev/instructions/ericandrechek/pacer/agents-md"><img src="https://agentmods.dev/badge/instructions/ericandrechek/pacer/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/ericandrechek/pacer/agents-md"><img src="https://agentmods.dev/badge/instructions/ericandrechek/pacer/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.17421 | $0.17421 |
| Opus 5.5 | $0.06968 | $0.06968 |
| Sonnet 5.5 | $0.03484 | $0.03484 |
| Haiku 4.5 | $0.01742 | $0.01742 |
Grade B, and why
Pacer 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 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- **Do not auto-write to `~/.claude/settings.json`** without explicit How it starts
The opening of the file, as written. The whole thing — 1,240 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent guide — Pacer
Native macOS Claude Code usage tracker. SwiftUI + SwiftData + Charts.
Single-binary menu-bar agent shape (LSUIElement=true). Two targets
(Pacer.app + PacerWidgets.appex) sharing data through an App Group.
See docs/design.md for the full v1 design.
Where to look first
-
docs/design.md— full architecture, data sources, schema, IPC, scope. -
docs/research/ccusage-reference.md— analysis ofccusageinternals: path discovery, JSONL schema, cost modes, dedup correctness, pricing source. Read before touching parsing or cost code: it is the tool Pacer was ported from, and most of its parsing rules still hold. -
docs/research/realtime-mechanisms.md— analysis of statusline, hooks, OTel, MCP for live Claude Code data. -
docs/research/tcc-app-management.md— investigation of the every-launch "would like to access data from other apps" prompt, what was tried, current signing/notarization state, and the open SMAppService verification question for v1 release. -
Ground truth is the raw JSONL, not ccusage (#299). Pacer's numbers are checked by its own independent passes: hand-computed fixtures (
JSONLScannerFixtureTestsfor tokens and what they cost,PriceLedgerTests,CuratedPriceHistoryTests),make test-ingest-parity(the line parser and the DuckDB importer must leave identical rows), the DuckDB differential gate (docs/duckdb-archive.md),make verify-data(every rollup against the raw samples of a real store), andPACER_COLD_START_PROBEon a frozen real~/.claude(correctness rule 8).ccusageis a reference: the opt-inCCusageGroundTruthTestscompare the token counts both tools count the same way (input, cache read, flat cache creation) againstdocs/research/ccusage-outputs/. Where Pacer differs, on purpose:What ccusage Pacer Why Cache writes one flat number 5m and 1h kept apart They are priced differently (rule 5). Streamed output first copy wins the finished copy wins First-wins keeps the mid-stream snapshot and under-counts output by ~63% (rule 7). If Pacer shows more output than ccusage, that is the fix working. Price today's catalog rate, for every day the rate in effect when the request was made A price change must not rewrite history ( PriceLedger).Long-context tier only the tokens above 200k at the higher rate the whole request at the tier rate once its prompt crosses the threshold That is how Anthropic bills it. Data residency ( inference_geo)not priced (per ccusage-reference.md)a multiplier on the standard rate Anthropic documents it as one. Claude Desktop not read its agent-mode sessions are counted too Desktop usage draws on the same limits.
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 Changed · +27 lines · +403 tokens per session fccf472b835a
- 5d ago Changed · +4 lines · +77 tokens per session 017ecbfc4fdf
- 6d ago Changed · +43 lines · +852 tokens per session 6e479dcd1455
- 7d ago Changed · +14 lines · +233 tokens per session 12231f2b071f
- 28d ago Changed · +6 lines · +95 tokens per session 7fe9fb2ded5f
- 1mo ago Changed · +26 lines · +350 tokens per session a616ee3d8be0
- 1mo ago Changed · +138 lines · +1,942 tokens per session 303dac582b32
- 1mo ago Changed · +46 lines · +691 tokens per session b3128b07b349
- 1mo ago Changed · +44 lines · +601 tokens per session 6f71b5117728
- 1mo ago First seen · 892 lines · 12,177 tokens per session scan B 9e482c9a2242
Pacer AGENTS.md is an instructions file published in the GitHub repository EricAndrechek/Pacer (11 stars, last pushed today), licensed MIT. It adds 17,421 tokens to every session, about $0.0697 per session on Opus 5.5. 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 2026-08-31.
Other instructions, from other repositories
vscode buildNext.instructions.md
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).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.