agentmods.dev

ai-first-toolkit AGENTS.md

A project guide for the TechWolf AI-First Toolkit, a repository that publishes add-ons for Claude Code, Codex, and Google Antigravity. It explains where each add-on lives and how installations are kept consistent across these tools.

In plain words
What is it for?
Use it when adding or renaming add-ons, updating marketplace files, preserving skill mappings, or installing the toolkit for Codex or Antigravity.
Why use it?
It prevents changes in one installation format from getting out of sync with the others.

Instructions file for CodexOpenCode

96 1mo ago B scan 649 tokens original MIT
View the source file techwolf-ai/ai-first-toolkit
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/techwolf-ai/ai-first-toolkit/agents-md
Clone the repo
git clone --depth 1 https://github.com/techwolf-ai/ai-first-toolkit

Made for: Codex, OpenCode.

Per session 649 This file is loaded in full into every session.
When invoked 649 The same file — it is already loaded in full.
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 $0.00649 $0.00649
Opus 5 $0.00324 $0.00324
Sonnet 5 $0.00130 $0.00130
Haiku 4.5 $0.00065 $0.00065

Measured 2d ago against content hash 460a290369fa, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

ai-first-toolkit AGENTS.md 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.

- Codex and Antigravity installs are both managed via `./install.sh` (Codex is the default; `--ide antigravity` installs one self-contained plugin dir per plugin into `~/.gemini/config/plugins/`, matching where Antigravi
AGENTS.md · 31 lines

How it starts

The opening of the file, as written. The whole thing — 31 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TechWolf AI-First Toolkit

This repository publishes TechWolf AI-first plugins and skills for Claude Code, Codex, and Google Antigravity.

Repository Purpose

  • plugins/<plugin>/ is the source of truth for each plugin.
  • The skills/ directory in each plugin holds the SKILL.md packages, shared verbatim across all three targets (agentskills.io spec). There is no per-target skill rewrite.
  • Per-target metadata is the only thing that differs:
    • Claude Code: .claude-plugin/marketplace.json (root) and plugins/<plugin>/.claude-plugin/plugin.json.
    • Antigravity / vendor-neutral: .agents/plugins/marketplace.json (root) and plugins/<plugin>/plugin.json (Antigravity plugin marker).
  • Codex and Antigravity installs are both managed via ./install.sh (Codex is the default; --ide antigravity installs one self-contained plugin dir per plugin into ~/.gemini/config/plugins/, matching where Antigravity keeps its own plugins).

Working Conventions

  • Preserve the mapping between a plugin and its installed skills across all targets.
  • When adding or renaming a plugin, update both marketplace manifests (.claude-plugin/marketplace.json and .agents/plugins/marketplace.json) and add the per-plugin plugin.json marker, keeping name/version/description in sync with .claude-plugin/plugin.json.
  • When changing a plugin, update Claude-, Codex-, and Antigravity-facing docs if behavior changes.
  • If install behavior changes, update README.md, plugin READMEs, and CHANGELOG.md.
  • Prefer adding plugin-specific agent guidance under plugins/<plugin>/codex/AGENTS.md (the installer copies it into the install state dir for both Codex and Antigravity targets).

Codex Install Notes

  • ./install.sh installs skills into ~/.codex/skills by default.
  • The installer also manages plugin metadata, verification, and uninstall state under ~/.codex/skills/.techwolf-ai-first/.
  • content-studio has a plugin-level Codex entry skill in addition to its specialized skills.
  • people-management has 8 skills that require /setup to be run first to configure org-specific frameworks and context.
  • knowledge-base ships three Python helper scripts (kb-index.py, kb-verify.py, kb-validate.py) that are copied into the user's project by /setup-knowledge-base. The plugin's /kb-answer workflow depends on kb-verify.py existing in the project's scripts/ directory.
  • session-tools has two skills: session-search (moved from ai-adoption in v1.9.0; host-aware, reads disk transcripts) and handoff (writes HANDOFF.md resume notes; no platform dependency). ai-adoption no longer includes session-search.
  • ai-adoption now covers two skills only: token-doctor and task-profile.

Read the full file on GitHub · 31 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 · 31 lines · 649 tokens per session scan B 460a290369fa

Subscribe to this mod's changes

ai-first-toolkit AGENTS.md is an instructions file published in the GitHub repository techwolf-ai/ai-first-toolkit (96 stars, last pushed 1mo ago), licensed MIT. It adds 649 tokens to every session, about $0.0032 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 2026-08-30.

agentskillsai-agentsai-codingai-firstai-toolsclaude-code
Related

Other instructions, from other repositories

thinkrail AGENTS.md

Instructions for JetBrains/thinkrail, covering thinkrail, module structure & boundaries (top-priority requirement), engine: pi only, in-process, architecture (three rings) and repo layout.

JetBrains/thinkrail · 4,503 tokens

snip CLAUDE.md

Instructions for edouard-claude/snip, covering claude.md, project overview, key concept, repository structure and architecture.

edouard-claude/snip · 2,038 tokens

wgm AGENTS.md

Instructions for agent-frontier/wgm, covering agents.md, build & run, validate (backpressure), operational notes and codebase patterns.

agent-frontier/wgm · 314 tokens

AmbyKit AGENTS.md

Instructions for ambystechcom/AmbyKit, covering agents.md — ambykit, what ambykit is, golden rules, project structure and build & test.

ambystechcom/AmbyKit · 1,216 tokens

coograph brutal-honesty.instructions.md

Global communication style — applies to all workspaces and conversations.

paullukic/coograph · 414 tokens

coograph testing.instructions.md

Conventions for test files — loaded only when editing tests.

paullukic/coograph · 397 tokens