aegis AGENTS.md

An AGENTS.md instruction file: guidance that coding agents read to understand how to install and use Aegis V2, an MCP server that reports findings about proposed file changes.

In plain words
What is it for?
Use it when adding Aegis V2 to a repository or telling an agent how to run its file-validation workflow.
Why use it?
It gives agents a prescribed setup sequence and explains that Aegis reports syntax, structure, security, dependency-cycle, and broken-caller findings without deciding whether a change should be blocked.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/wei9072/aegis/agents-md
Clone the repo
git clone --depth 1 https://github.com/wei9072/aegis

Made for: Codex, OpenCode.

Per session 2,698 This file is loaded in full into every session.
When invoked 2,698 The same file — it is already loaded in full.
Security scan C 2 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.02698 $0.02698
Opus 5 $0.01349 $0.01349
Sonnet 5 $0.00540 $0.00540
Haiku 4.5 $0.00270 $0.00270

Measured 2d ago against content hash 27e14830a6f3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

aegis AGENTS.md scanned grade C with 2 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.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

cargo --version || curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && source "$HOME/.cargo/env"

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

cargo --version || curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && source "$HOME/.cargo/env"
AGENTS.md · 263 lines

How it starts

The opening of the file, as written. The whole thing — 263 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Instructions for AI coding agents (Claude Code, Cursor, Aider, etc.)

This file tells you, the AI agent, how to help a user adopt Aegis V2 in their project. The user almost certainly typed something like "add Aegis to my repo". Follow these commands verbatim; don't paraphrase, don't skip.

If anything in this file conflicts with your interpretation of README.md, follow this file. README is for human evaluators; this is for you.


What Aegis V2 is, in one paragraph

A judgment-free MCP server for LLM-driven coding workflows. The single tool, validate_file, returns structured findings about a proposed file write — syntax errors, structural-smell deltas, security pattern matches, cross-file cycles, broken-caller lists. Aegis does not block writes, does not rank findings, does not say "this is bad". You — the consuming agent — decide which findings to act on. Aegis only ensures the data is on the table.

Full design context: README.md. Read that when the user asks "what does this thing do".


Setup — the canonical install sequence

V2 ships a single binary: aegis-mcp. Run these from the user's home or workspace dir.

# 0. Prerequisites
git --version
cargo --version || curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && source "$HOME/.cargo/env"

# 1. Clone, enter
git clone https://github.com/wei9072/aegis ~/aegis
cd ~/aegis

# 2. Build + install. ~1-2 min first time.
cargo install --path crates/aegis-mcp

# 3. Verify (do not skip)
which aegis-mcp           # confirms $PATH installation
aegis-mcp --version 2>&1 | head -1 || echo "(aegis-mcp does not have --version; absence of 'command not found' is the success signal)"

After step 3 succeeds, aegis-mcp is ready to be configured as an MCP server in your client.

Common breakage:

  • cargo: command not found → restart shell or source "$HOME/.cargo/env".
  • failed to compile aegis-corerustup update (need 1.74+).

Integration — configure the MCP client

Read the full file on GitHub · 263 lines

Changes

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.

  1. 2d ago First seen · 263 lines · 2,698 tokens per session scan C 27e14830a6f3

Subscribe to this mod's changes

aegis AGENTS.md is an instructions file published in the GitHub repository wei9072/aegis (1 stars, last pushed 3mo ago), licensed MIT. It adds 2,698 tokens to every session, about $0.0135 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.