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/skwallace36/pepper/claude-mdgit clone --depth 1 https://github.com/skwallace36/PepperWrote 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/skwallace36/pepper/claude-md)<a href="https://agentmods.dev/instructions/skwallace36/pepper/claude-md"><img src="https://agentmods.dev/badge/instructions/skwallace36/pepper/claude-md.svg" alt="Measured on agentmods" 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 | $0.01598 | $0.01598 |
| Opus 5 | $0.00799 | $0.00799 |
| Sonnet 5 | $0.00320 | $0.00320 |
| Haiku 4.5 | $0.00160 | $0.00160 |
Grade A, and why
Pepper CLAUDE.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 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.
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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
pepper
Pepper is a dylib injected into iOS simulator apps via DYLD_INSERT_LIBRARIES. It starts a WebSocket server inside the app process; MCP server and CLI tools connect to it. No source patches needed.
Source of truth for dylib code is dylib/. Config via .env (see .env.example). Run make help for all targets.
Adapters are optional app-specific modules (deep link routes, icon mappings, custom tools) compiled alongside the dylib. Set APP_ADAPTER_TYPE and ADAPTER_PATH in .env. Without an adapter, Pepper runs in generic mode.
MCP Tools
Tool definitions live in pepper_ios/mcp_tools_*.py modules. .mcp.json configures how to launch the MCP server; tools/pepper-mcp is the entry script. app_look is the primary observation tool. It sends the introspect command with mode=map.
app_look, app_build, app_build_hw, app_status, app_record, app_snapshot, app_console, app_network, app_debug, app_perf, app_swiftui, app_automation, app_eval, ui_tap, ui_scroll, ui_swipe, ui_input, ui_toggle, ui_gesture, ui_query, ui_accessibility, nav_go, nav_back, nav_dismiss, nav_keyboard, nav_dialog, nav_screen, state_vars, state_tools, net_tools, sys_tools, sim_control, sim_raw
Tools use a two-word prefix convention for routing. Standalone tools are high-frequency; grouped tools bundle related subcommands under one tool with a command parameter.
Standalone (23):
- app_: app_look, app_build, app_build_hw, app_status, app_record, app_snapshot, app_console, app_network, app_eval
- ui_: ui_tap, ui_scroll, ui_swipe, ui_input, ui_toggle, ui_gesture
- nav_: nav_go, nav_back, nav_dismiss, nav_keyboard, nav_dialog, nav_screen
- state_: state_vars
- sim_: sim_control, sim_raw
Grouped (9):
- ui_query (find, tree, verify, read, assert)
- ui_accessibility (audit, action, events)
- app_debug (layers, crash_log, constraints, responder_chain, target_actions, highlight, lifecycle, notifications, webview)
- app_perf (perf, animations, heap, hangs, renders, timers)
- app_swiftui (renders, body)
- app_automation (script, wait_for, wait_idle, concurrency)
- state_tools (defaults, keychain, clipboard, cookies, sandbox, coredata, undo_manager, flags)
- net_tools (mock, simulate, http_call, timeline)
- sys_tools (push, orientation, locale, appearance, dynamic_type, hook, frameworks, usage)
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 First seen · 83 lines · 1,598 tokens per session scan A 709666ac42cd
Pepper CLAUDE.md is an instructions file published in the GitHub repository skwallace36/Pepper (103 stars, last pushed 4mo ago), licensed MIT. It adds 1,598 tokens to every session, about $0.0080 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
ios-swiftui-accessibility-techniques CLAUDE.md
Claude Code instructions for cvs-health/ios-swiftui-accessibility-techniques, covering project rules, copyright header (required), rules, swift files and markdown files.
MarkdownView AGENTS.md
Instructions for keitaoouchi/MarkdownView, covering agents, technical components, relationships (data/event flow overview), extension points (overview) and platform / distribution.
AniShelf AGENTS.md
Instructions for samuelhe52/AniShelf, covering anishelf - developer notes, workflow, code style, testing and commits.
swift-architecture-skill CLAUDE.md
Instructions for efremidze/swift-architecture-skill, covering swift architecture skill, project structure, how the skill works and key conventions.
SwiftUIShaders AGENTS.md
Instructions for krispuckett/SwiftUIShaders, covering agents.md: architecture for coding agents, what this package is, how swiftui metal shader effects work (the mental model), how to add a new shader and how to modify an existing shader.
swift-architecture-skill AGENTS.md
Instructions for efremidze/swift-architecture-skill, covering agents guidelines for swift architecture skill, working with this repository, 1. understanding the skill, 2. modifying architecture playbooks and 3. testing and validation.