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/ahad690/open-app-intel/agents-mdgit clone --depth 1 https://github.com/Ahad690/open-app-intelWrote 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/ahad690/open-app-intel/agents-md)<a href="https://agentmods.dev/instructions/ahad690/open-app-intel/agents-md"><img src="https://agentmods.dev/badge/instructions/ahad690/open-app-intel/agents-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 | $0.00582 | $0.00582 |
| Opus 5 | $0.00291 | $0.00291 |
| Sonnet 5 | $0.00116 | $0.00116 |
| Haiku 4.5 | $0.00058 | $0.00058 |
Grade A, and why
open-app-intel 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 3d 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 — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — running AppScope from any agentic CLI
AppScope is a Claude Code skill + local MCP server, but CLI-agnostic: the
intelligence lives in skills/appscope/SKILL.md plus deterministic Python (appscope.*
modules), so any agentic coding CLI (Claude Code, Codex, OpenCode, Cursor, Gemini CLI,
Copilot CLI, Qwen, Kimi, Grok) can drive it. This file is the entry point those tools read.
The rules you may never break
- P1 — Every number carries confidence + method + provenance. Envelope:
{value, low, high, confidence, method, sources, flags}. - P2 — Estimates are ranges, capped at MEDIUM. HIGH is reserved for directly observed facts (a captured rank, a real install bucket, a real review count). A modeled estimate is never HIGH.
- P3 — Proxies, not dollars, for ads. The ad module emits intensity proxies and a disclaimer; it never outputs USD spend.
- N4 — Free-app revenue is never invented. Returns not estimable without an ARPU.
- P8 — Local-first; federate only public anchors. Ads and creator data never leave
the machine;
assert_public_onlyaborts any contribution carrying ad/creator/identity.
If a segment is uncalibrated, the estimator returns NONE / uncalibrated — say so. It is correct for the tool to refuse rather than guess.
How to run it
- Read
skills/appscope/SKILL.md— the full operating procedure. - Drive the JSON CLI:
python -m appscope.cli collect --app <id> [--charts] # append-only observations
python -m appscope.cli summary --app <id> # observed facts (HIGH)
python -m appscope.cli estimate --app <id> # banded estimate (≤ MEDIUM)
python -m appscope.cli report --app <id> # the HTML deliverable
python -m appscope.cli backup # timestamped DB snapshot
- Or point a local MCP client at
python -m appscope.mcp_server(tools:app_estimate,ad_intensity,creator_mentions,rank_history).
Contract enforcement
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.
- 3d ago First seen · 45 lines · 582 tokens per session scan A faf6b38a3ee1
open-app-intel AGENTS.md is an instructions file published in the GitHub repository Ahad690/open-app-intel (0 stars, last pushed today), licensed MIT. It adds 582 tokens to every session, about $0.0029 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 2026-08-31.
Other instructions, from other repositories
claude-code-apple-skills CLAUDE.md
Claude Code instructions for rshankras/claude-code-apple-skills, covering claude.md, repository purpose, architecture, skill structure and yaml frontmatter (required).
shotluma AGENTS.md
Instructions for realZachi/shotluma, covering shotluma agent instructions, required workflow, commands, non-negotiable rules and rulebook.
storeshots-mcp AGENTS.md
Instructions for mtbun/storeshots-mcp, covering agents.md, what this project is, architecture, non-negotiable principles and layout contract (compose).
app-store-operator CLAUDE.md
Instructions for meyusufdemirci/app-store-operator, covering app store operator — mcp server, codebase, server wiring, tool module contract and prompts.
appshot AGENTS.md
Instructions for trunghaiy/appshot, covering agents.md, repository overview, architecture, repository structure and agent skills specification.
appshot CLAUDE.md
Instructions for trunghaiy/appshot, covering appshot — claude code project brief, key constraint, where things live, build and git.