agent-memory-atlas: Skill for Claude Code

.agents/skills/screen-repository/SKILL.md

screen-repository is a skill for Claude Code, Codex from neoneye/agent-memory-atlas. It costs 54 tokens per session (2,783 once invoked), scanned C, original, MIT.

A safety check for a downloaded code repository before an agent reads or runs it. It looks for automatic hooks and dependencies whose versions are not fixed, both of which can execute unwanted code on a developer’s computer.

In plain words
What is it for?
Use it on cloned repositories before adding or reanalyzing a memory system, especially when the checkout comes from an unfamiliar source.
Why use it?
Opening, building, or installing a repository can run code before its contents are fully inspected. Screening helps identify that supply-chain and checkout risk before analysis begins.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: reads .claude/ paths; mentions CLAUDE.md; installed under .agents/ (shared by several agents).

This is neoneye/agent-memory-atlas's own configuration. It tells Claude Code and Codex how to work on agent-memory-atlas itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything agent-memory-atlas configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is python3 scripts/screen_repo.py /path/to/checkout.

Reuse

Borrowing it

Nothing to install: this file belongs to neoneye/agent-memory-atlas. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/neoneye/agent-memory-atlas/main/.agents/skills/screen-repository/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/neoneye/agent-memory-atlas

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for screen-repository

README.md
[![agentmods](https://agentmods.dev/badge/skills/neoneye/agent-memory-atlas/screen-repository/github.svg)](https://agentmods.dev/skills/neoneye/agent-memory-atlas/screen-repository)
Your own site
<a href="https://agentmods.dev/skills/neoneye/agent-memory-atlas/screen-repository"><img src="https://agentmods.dev/badge/skills/neoneye/agent-memory-atlas/screen-repository/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for screen-repository

Your own site · 80×15
<a href="https://agentmods.dev/skills/neoneye/agent-memory-atlas/screen-repository"><img src="https://agentmods.dev/badge/skills/neoneye/agent-memory-atlas/screen-repository.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 54 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,783 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 2 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00054 $0.02783
Opus 5 $0.00027 $0.01392
Sonnet 5 $0.00011 $0.00557
Haiku 4.5 $0.00005 $0.00278

Measured 10d ago against content hash d7947cf25783, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade C, and why

screen-repository 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 10d 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.

a shell (`curl … | sh`).

Makes network callslowCapability

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

a shell (`curl … | sh`).
.agents/skills/screen-repository/SKILL.md · 233 lines

How it starts

The opening of the file, as written. The whole thing — 233 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Screen Repository

Run this before add-memory-system and before reanalyze-memory-system, every time, including on a repository already in the atlas. A re-read clones a newer commit than the one that was screened, and the newest commit is exactly where a compromise would arrive.

The threat model, stated plainly

This atlas is produced by cloning strangers' repositories onto a personal machine and frequently running their build, their tests or their demo. The risk is not what a project does to its own users. It is what a checkout does to this laptop, between git clone and the first read.

Two shapes, failing differently:

Auto-executing configuration needs no command typed. An agent-harness hook firing on session start or before a tool call; a devcontainer postCreateCommand; a .envrc that direnv sources on cd; a VS Code task with runOn: folderOpen; a .gitattributes smudge filter that runs during checkout itself. Opening or entering the directory is enough.

Unpinned dependencies are the 2026 supply-chain shape. A manifest names a range, a maintainer account is compromised, a new version publishes, and every machine that installs afterwards executes the attacker's postinstall. The compromise is not in the repository being read, and reading the repository will never reveal it.

Pinning does not make a dependency safe. It makes it the same thing today as when someone last looked, which is the only property that can be checked from a clone.

Step 1 — clone without executing anything

git clone --no-checkout --recurse-submodules=no <url> <dir>
cd <dir> && git checkout <ref>

--recurse-submodules=no matters: a submodule pulls another untrusted tree, and the screen below only sees the outer one. If .gitmodules exists, screen each submodule separately or leave it uninitialised.

Do not cd into the directory in a shell where direnv is installed until after step 2, and do not open it in an editor.

Step 2 — run the screen

Read the full file on GitHub · 233 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. 10d ago First seen · 233 lines · 54 tokens per session scan C d7947cf25783

Subscribe to this mod's changes

screen-repository is a skill published in the GitHub repository neoneye/agent-memory-atlas (64 stars, last pushed today), licensed MIT. It adds 54 tokens to every session and 2,783 once invoked, about $0.0003 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 2026-08-30.

Related

Other skills, from other repositories

thoughtline-memory

Persistent, project-aware memory for AI coding agents. Use thoughtline whenever you make a decision, fix a bug, learn something non-obvious, or need to recall prior work.

AgusLoza2021/Thoughtline · 40 tokens

Effective Memory

The essential habits for an AI agent with memory — session bookends, learning triggers, verification, safety, and the operational discipline that turns raw recall into compounding intelligence. Pinned, always-injected.

plur-ai/plur · 44 tokens

recall

Recall this repository's OwnMem local memory before changing code, and keep it healthy. Use when a repository contains .ownmem/, when past debugging lessons could apply ("have we hit this before", "why is it done this way"), or when the user mentions ownmem, project memory, or recalling across sessions.

grpcer/ownmem · 66 tokens

init

Install or update OwnMem in the current repository. Use when the user asks to set up OwnMem, add local project memory for coding agents, or refresh an existing OwnMem installation after a version bump.

grpcer/ownmem · 43 tokens

honcho-integration

Integrate Honcho memory into existing Python or TypeScript codebases. Use when adding Honcho SDK, setting up peers, configuring sessions, and accessing Honcho's representation.

plastic-labs/honcho · 40 tokens

honcho-memory

Concepts and strategy for using a connected Honcho as persistent memory of the user — the recall/record loop and session and peer design. Start here to understand how Honcho memory works, then connect — via a first-class integration for your environment if one exists (preferred), or raw MCP tools (covered here) or the…

plastic-labs/honcho · 91 tokens