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/salavender/antigravity-compound-engineering-plugin/gemini-mdgit clone --depth 1 https://github.com/salavender/antigravity-compound-engineering-pluginWrote 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/salavender/antigravity-compound-engineering-plugin/gemini-md)<a href="https://agentmods.dev/instructions/salavender/antigravity-compound-engineering-plugin/gemini-md"><img src="https://agentmods.dev/badge/instructions/salavender/antigravity-compound-engineering-plugin/gemini-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 | $0.01242 | $0.01242 |
| Opus 5 | $0.00621 | $0.00621 |
| Sonnet 5 | $0.00248 | $0.00248 |
| Haiku 4.5 | $0.00124 | $0.00124 |
Grade A, and why
antigravity-compound-engineering-plugin GEMINI.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 5d 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 — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Compound Project - Agent Configuration
Project Overview
This project implements the Compound Learning System for {PROJECT_NAME} - a self-improving knowledge system based on the "Compounding Engineering" philosophy.
Core Principle
Each unit of engineering work should make subsequent units of work easier—not harder.
Workflows Available
Full index: See .agent/workflows/README.md for all commands and quick start guide.
Use these commands for systematic development:
| Command | When |
|---|---|
/explore |
Deep investigation before planning |
/specs |
Before multi-week initiatives |
/plan |
Before starting significant work |
/work |
Execute plans systematically |
/review |
Before merging, self-review |
/compound |
After solving problems ("that worked!") |
/housekeeping |
Before git push (cleanup & archive) |
/tracker-routing |
Before creating or moving work between Linear and GitHub |
Knowledge Persistence
Solutions are documented in docs/solutions/ and explorations in docs/explorations/ with:
- YAML frontmatter for searchability
- Categorized by problem type
- Schema validated (
schema.yaml)
Before solving a problem: Search docs/solutions/ and docs/explorations/ for prior knowledge.
After solving a problem: Run /compound to document it.
Compounding Loop
/explore (optional) → /specs (large) → /plan (per phase) → /work → /review → /compound → /housekeeping → repeat
Work-surface boundary
The compound system is not a second project-management database.
Git docs/specs/ADRs = durable truth
GitHub Issues/PRs = engineering execution + evidence
Linear = temporary attention, brainstorming, decisions, gates
Use /tracker-routing before creating or moving work objects.
Linear
Use Linear for collaborative reasoning, product/strategy questions, architecture alternatives, risk acceptance, go/no-go decisions, and operator/human attention that genuinely needs a shared surface.
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.
- 5d ago First seen · 110 lines · 1,242 tokens per session scan A 4f9572a3858e
antigravity-compound-engineering-plugin GEMINI.md is an instructions file published in the GitHub repository salavender/antigravity-compound-engineering-plugin (47 stars, last pushed 21d ago), licensed MIT. It adds 1,242 tokens to every session, about $0.0062 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
aeon CLAUDE.md
Instructions for aeonfun/aeon, covering aeon, how aeon works, strategy, voice and soul file hierarchy (read in this order).
cognirepo CLAUDE.md
Claude Code instructions for ashlesh-t/cognirepo, covering claude.md, key rules, session start sequence (run in this order), behavioral confirmation rule and personas (cognirepo-402, cognirepo-403).
wayland-core copilot-instructions.md
Copilot instructions for FerroxLabs/wayland-core, covering ijfw rules, output discipline, memory routing, context discipline and cross-audit.
mcp-structured-memory CLAUDE.md
Claude Code instructions for nmeierpolys/mcp-structured-memory, a project described as: Structured Memory MCP Server.
inkwell-memory CLAUDE.md
Instructions for veronchenko/inkwell-memory, covering claude.md — inkwellmemory, layout, multi-tenant mode (inkwellmultitenant=1), conventions and testing.
RNR-Enhanced-Cognee AGENTS.md
AGENTS.md instructions for vincentspereira/RNR-Enhanced-Cognee, covering rnr enhanced cognee implementation for codex, critical requirements, 1. ascii-only output (no unicode encoding), 2. dynamic categories (no hardcoded categories) and 3. standard memory mcp interface.