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/fmarzochi/egc/copilot-instructionsgit clone --depth 1 https://github.com/Fmarzochi/EGCWrote 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/fmarzochi/egc/copilot-instructions)<a href="https://agentmods.dev/instructions/fmarzochi/egc/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/fmarzochi/egc/copilot-instructions.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 | $0.00512 | $0.00512 |
| Opus 5 | $0.00256 | $0.00256 |
| Sonnet 5 | $0.00102 | $0.00102 |
| Haiku 4.5 | $0.00051 | $0.00051 |
Grade A, and why
EGC copilot-instructions.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 — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Copilot instructions for EGC
EGC is a local-first MCP runtime that gives every AI coding agent a shared persistent memory (egc-memory), a safety layer that validates commands and writes before they run (egc-guardian), and shell-output compression (Token Crusher). It installs into 20 AI coding tools from one npm package (@egchq/egc). Memory lives encrypted in ~/.egc on the user's machine and must never reach the repository.
Repository conventions
- All repository content is written in English, except the files under translations/.
- Flag any populated memory content in AGENTS.md, GEMINI.md, .cursor/rules/egc-context.mdc or .trae/rules/egc-context.md: those propagation files must ship with an empty structure only, and CI enforces it.
README and translations
- README.md must keep the exact catalog sentence matching "access to N agents, N skills, and N commands" and the "Works natively with ..." provider sentence: CI validates both.
- The 7 translations (ar, es, hi, ja, ko, pt, ru) must mirror the heading levels of README.md in the same order; adding or removing a section in one language only is a defect.
- The Support EGC section and everything below it (sponsors, backers, OpenSSF footer) changes only when the maintainer edits it deliberately.
Code review priorities
- Commits need a DCO Signed-off-by line and follow conventional commits.
- New code never uses Math.random for anything security-adjacent: use crypto.randomInt (SonarCloud S2245 gates the merge).
- Binaries invoked from scripts are resolved from fixed paths with a PATH fallback, not bare names (S4036 pattern used across scripts/).
- Any change touching files shared across concurrent EGC processes (~/.egc state, encryption key, install-state, lockfiles) requires a concurrent-access regression test; check for read-merge-write races and lock steal in upserts.
- Shell-facing features must be fail-open: a broken hook or missing binary must never block the user's command.
- Version bumps happen only in dedicated release PRs that update the whole release surface (11 files); flag any stray version change elsewhere.
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 · 50 lines · 512 tokens per session scan A 884d6314f20e
EGC copilot-instructions.md is an instructions file published in the GitHub repository Fmarzochi/EGC (48 stars, last pushed today), licensed Apache-2.0. It adds 512 tokens to every session, about $0.0026 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
brain-os AGENTS.md
Instructions for brainOS-HQ/brain-os, covering agents.md, context, the hard rule, tool routing (in order) and mutation safety.
skills AGENTS.md
AGENTS.md instructions for Yesterday-AI/skills, covering agents.md, what this repo is, hard rules (do not violate), common tasks and project memory.
vibe-mcp AGENTS.md
Instructions for VibeTechnologies/vibe-mcp, covering vibe-mcp agent memory, canonical eval reference, relay and connection semantics, what counts as real e2e and fake extension test harness rules.
kleosrules AGENTS.md
Instructions for kleosr/kleosrules, a project described as: Cursor harness pack: user rules, skills, Bash hooks, local HANDOFF memory. macOS, Linux, Windows (WSL).
knowit AGENTS.md
Instructions for ismaelkedir/knowit: This project uses Knowit as the default persistent knowledge base for AI coding agents.
memex CLAUDE.md
Instructions for iamtouchskyer/memex, covering memex — agent working instructions, what is this, before you touch anything, how the code is layered and files you'll actually need.