setup-multi-agent

A project setup tool that adds shared AGENTS.md and CLAUDE.md instruction files for Claude Code, Codex CLI, and Antigravity CLI. These are command-line AI coding tools, and the files give them common project guidance.

In plain words
What is it for?
Use it when starting or standardizing a project across those three coding CLIs, including updating its ignore-file settings.
Why use it?
It prevents each coding tool from using a different copy of the project's instructions.

Skill for Claude CodeCodex

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 skills/fuzzykala/cc-templates/setup-multi-agent
Any agent
npx skills add FuzzyKala/cc-templates --skill setup-multi-agent
Clone the repo
git clone --depth 1 https://github.com/FuzzyKala/cc-templates

Made for: Claude Code, Codex.

Per session 79 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,857 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00079 $0.01857
Opus 5 $0.00039 $0.00928
Sonnet 5 $0.00016 $0.00371
Haiku 4.5 $0.00008 $0.00186

Measured 2d ago against content hash e6dc9ca4a248, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

setup-multi-agent scanned grade B with 1 finding 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

- If it exists, append `${TEMPLATES_DIR}/gitignore-additions.txt` to it — but first grep for `.gemini/` to avoid double-appending if the user runs the skill twice.
.agents/skills/setup-multi-agent/SKILL.md · 128 lines

How it starts

The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/setup-multi-agent — Bootstrap a new project

Drop the AGENTS.md + CLAUDE.md pair into a target project so all three coding CLIs (Claude Code, Codex CLI, Antigravity CLI) load the same canonical project context. AGENTS.md is the canonical source; CLAUDE.md is a 1-line @AGENTS.md import so Claude Code (which doesn't yet natively read AGENTS.md) gets the same content.

Prerequisites

The cc-templates repo must be available locally. The skill reads three files from its templates/ directory:

  • templates/AGENTS.md.template
  • templates/CLAUDE.md.template
  • templates/gitignore-additions.txt

Step 0 below locates that directory. If any of the three files is missing after resolution, fail loudly and stop.

Steps

0. Locate the templates directory

This skill is usually invoked from a symlinked install at the skill directory (via symlink chain) that points back into the cct repo, so templates/ is NOT a sibling of SKILL.md. Resolve the real cct repo root by following the skill's own symlink and walking up three levels.

The harness reports the skill's base directory at invocation. Pass that path through readlink -f and ascend three directories — the skill lives at <cct-repo>/.agents/skills/setup-multi-agent/, so up-three lands at <cct-repo>:

# Replace <skill-base-dir> with the path the harness reported at invocation.
SKILL_DIR_RESOLVED="$(readlink -f "<skill-base-dir>")"
CCT_ROOT="$(readlink -f "${SKILL_DIR_RESOLVED}/../../..")"
TEMPLATES_DIR="${CCT_ROOT}/templates"

# Verify the three template files exist; fail loudly if any are missing.
missing=0
for f in AGENTS.md.template CLAUDE.md.template gitignore-additions.txt; do
  if [ ! -r "${TEMPLATES_DIR}/${f}" ]; then
    echo "ERROR: Missing template at ${TEMPLATES_DIR}/${f}" >&2
    missing=1
  fi
done
[ "${missing}" -eq 0 ] || exit 1

echo "Templates dir resolved: ${TEMPLATES_DIR}"

Use the resolved ${TEMPLATES_DIR} (or the absolute path the snippet prints) in all subsequent steps. This works whether the skill is invoked from a symlinked install or directly from inside the cct repo; the readlink -f step is what lets symlinked installs resolve to the real cct repo root instead of the symlink path.

Read the full file on GitHub · 128 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 · 128 lines · 79 tokens per session scan B e6dc9ca4a248

Subscribe to this mod's changes

setup-multi-agent is a skill published in the GitHub repository FuzzyKala/cc-templates (1 stars, last pushed 20d ago), licensed MIT. It adds 79 tokens to every session and 1,857 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.