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/ericandrechek/pacer/agents-mdgit clone --depth 1 https://github.com/EricAndrechek/PacerWhat 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 | $0.10285 | $0.10285 |
| Opus 5 | $0.05143 | $0.05143 |
| Sonnet 5 | $0.02057 | $0.02057 |
| Haiku 4.5 | $0.01029 | $0.01029 |
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 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.
- **Do not auto-write to `~/.claude/settings.json`** without explicit How it starts
The opening of the file, as written. The whole thing — 749 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— ground-truth analysis ofccusageinternals: path discovery, JSONL schema, cost modes, dedup correctness, pricing source. Read before touching parsing or cost code.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.docs/research/ccusage-outputs/— capturedbun x ccusageJSON outputs for the local dataset. Use these as ground-truth in tests — every metric Pacer surfaces should matchccusage's number for the same range, with exactly two deliberate deviations:- the cache 5m/1h split (we track them separately, ccusage flattens);
- output tokens on streamed messages — ccusage dedups first-wins, which keeps the mid-stream snapshot instead of the finished message and under-counts output by ~63% on a real corpus (correctness rule §7). Pacer is deliberately higher here. If a ccusage comparison shows us reporting more output than ccusage, that is the fix working — do not "correct" it back.
AGENTS.md→ "Performance — invariants and patterns" (below) — read before adding ANY@Query,FetchDescriptor, computed view property, widget provider, or new rollup table. Codifies hard-won rules from five rounds of read-path optimization. The rules look nitpicky in isolation; in aggregate they're what keeps the app responsive while the in-process scan loop is firing every 5–60s.docs/perf-tuning.md— current cycle-time / CPU state, the measurement tooling (phase-timed scan log,make perf-snapshot), every perf commit's mechanism + measured win, and the open refactors that are deferred. Read before reintroducing animations, per-cycle SwiftData fetches, or adding any new always-running background work.
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 · 749 lines · 10,285 tokens per session scan B 4e15019f2364
Pacer AGENTS.md is an instructions file published in the GitHub repository EricAndrechek/Pacer (8 stars, last pushed 21d ago), licensed MIT. It adds 10,285 tokens to every session, about $0.0514 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 2026-08-31.
Other instructions, from other repositories
barkeep AGENTS.md
Instructions for iannuttall/barkeep, covering agent notes, product rules, repo map, commands and app structure.
openquack AGENTS.md
Instructions for larryxiao/openquack, covering agents.md, posture, before you start, workflow and pr template (the required shape).
codex-reset-watcher AGENTS.md
Instructions for jordan-edai/codex-reset-watcher, covering agent notes, current state, product decisions, git and workspace rules and verification commands.
Tokfuel AGENTS.md
Instructions for Tokfuel/Tokfuel, covering agents.md（ai セッションの作業合意）, これは何か, project structure, グラウンドルール（違反禁止） and 検証ゲート.
CopyCat CLAUDE.md
Instructions for andyhtran/CopyCat, covering copycat, releasing and icons.
LanGuard CLAUDE.md
Instructions for roypadina/LanGuard, covering languard, layout, core behaviour (edge-based), build / test / run and build (xcodebuildmcp session default = this workspace).