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/milisp/agent-insights/agents-mdgit clone --depth 1 https://github.com/milisp/agent-insightsWhat 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 | $0.00757 | $0.00757 |
| Opus 5 | $0.00378 | $0.00378 |
| Sonnet 5 | $0.00151 | $0.00151 |
| Haiku 4.5 | $0.00076 | $0.00076 |
Grade A, and why
agent-insights 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 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.
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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Insights Backend
Rust backend API for scanning and aggregating AI agent usage data.
Features
- Scans local agent directories (Claude, Gemini, Codex)
- Parses JSON/JSONL session files
- Aggregates data by date for heatmap visualization
- RESTful API with CORS support
Architecture
agent-insights/
├── src/
│ ├── domain/ # Domain models
│ │ ├── record.rs # AgentRecord, TokenInfo
│ │ └── heatmap.rs # HeatmapData, DayActivity
│ ├── scanner/ # File scanning
│ │ └── mod.rs # FileScanner with walkdir
│ ├── agents/ # Agent-specific parsers
│ │ ├── claude.rs # JSONL parser for ~/.claude/projects
│ │ ├── gemini.rs # JSON parser for ~/.gemini/tmp
│ │ └── codex.rs # Scanner for ~/.codex
│ ├── services/ # Business logic
│ │ ├── collect.rs # CollectionService
│ │ └── aggregate.rs # AggregationService
│ ├── api/ # HTTP API
│ │ └── heatmap.rs # Heatmap endpoints
│ └── main.rs # Axum server
│ └── ui/ # frontend (react + vite + tailwind + shadcn + bun)
API Endpoints
GET /api/heatmaps
Returns all agent heatmaps
{
"Claude": {
"agent": "Claude",
"data": [
{"date": "2025-12-02", "count": 20, "size": 5096527},
{"date": "2025-12-03", "count": 27, "size": 1687467}
],
"max_count": 82,
"total_files": 666,
"total_size": 123456789
},
"Gemini": {...},
"Codex": {...}
}
GET /api/heatmap/:agent
Returns heatmap for specific agent (claude, gemini, codex)
{
"agent": "Claude",
"data": [...],
"max_count": 82,
"total_files": 666,
"total_size": 123456789
}
Running
cargo run
Server starts on http://127.0.0.1:3001
Implementation Details
Claude Scanner
- Location:
~/.claude/projects/ - Format: JSONL (newline-delimited JSON)
- Extracts: session_id, token usage (input/output/cached)
- Pattern: One event per line
Gemini Scanner
- Location:
~/.gemini/tmp/{hash}/chats/ - Format: JSON
- Extracts: session_id, token usage
- Pattern: Multiple chat files per session
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 · 102 lines · 757 tokens per session scan A 1b14f9eb22af
agent-insights AGENTS.md is an instructions file published in the GitHub repository milisp/agent-insights (2 stars, last pushed 5mo ago), licensed MIT. It adds 757 tokens to every session, about $0.0038 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
obsidian-agent-client AGENTS.md
AGENTS.md instructions for RAIT-09/obsidian-agent-client, covering agent client plugin - llm developer guide, architecture, data flow, acp event flow (single path) and permission flow.
memex GEMINI.md
Instructions for STiFLeR7/memex, covering memex, agent instructions, execution mode, non-negotiables and what is in scope.
kmux AGENTS.md
Instructions for kkd927/kmux, covering agents.md, kmux focus and test discipline.
obsidian-llm-hub AGENTS.md
Instructions for takeshy/obsidian-llm-hub, a project described as: Obsidian plugin for AI chat, workflow automation, and semantic search — supports Gemini, OpenAI, OpenRouter, Grok, Ollama, and CLI backends.
docker-gemini-cli GEMINI.md
Instructions for naoyoshinori/docker-gemini-cli, covering 1. core functionality, 2. project structure and key files, 2.1. image variants, 2.2. automation and 3. documentation.
gemini-faf-mcp GEMINI.md
Instructions for Wolfe-Jam/gemini-faf-mcp, covering gemini.md — gemini-faf-mcp, test & verify, where things live, stack and before changing things.