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/heapy/kortex/agents-mdgit clone --depth 1 https://github.com/Heapy/kortexWhat 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 | $0.01306 | $0.01306 |
| Opus 5 | $0.00653 | $0.00653 |
| Sonnet 5 | $0.00261 | $0.00261 |
| Haiku 4.5 | $0.00131 | $0.00131 |
Grade A, and why
kortex AGENTS.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 yesterday.
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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
kortex
One tree, three hosts. The same skills ship as Claude Code plugins, Codex plugins, and Junie extensions, so every change has to be made in each host's manifest, not just the one you are looking at.
Layout
| Path | What it is |
|---|---|
plugins/<plugin>/skills/<skill>/SKILL.md |
the skill itself: YAML frontmatter (name, description) then the body |
plugins/<plugin>/agents/<agent>.md |
a subagent: YAML frontmatter then the prompt; loaded by Claude Code and Junie |
plugins/<plugin>/plugin.json |
portable Agent Plugins manifest; no host reads it yet |
plugins/<plugin>/.claude-plugin/plugin.json |
Claude Code manifest |
plugins/<plugin>/.codex-plugin/plugin.json |
Codex manifest |
plugins/<plugin>/extension.json |
Junie manifest |
.claude-plugin/marketplace.json |
Claude Code marketplace index |
.junie-extension/marketplace.json |
Junie marketplace index |
scripts/sync-agents.main.kts |
regenerates the Codex copy of every agent |
docs/ |
field reports; not shipped with any plugin |
A skill's description is what the host matches a user request against — write it as the set of
phrases that should trigger it, not as a summary.
Specs
Two specifications govern the tree, and they split cleanly: one owns the skill, the other owns the directory around it.
| Spec | Owns | Where |
|---|---|---|
| Agent Skills | SKILL.md: frontmatter fields, name/description limits, scripts/, references/, assets/ |
https://agentskills.io/specification |
| Agent Plugins 1.0.0 | the plugin directory: plugin.json in the plugin root, skills discovered at skills/, MCP at mcp.json |
https://github.com/agentplugins/agent-plugins-spec, spec/1.0.0.md, plugin.schema.json |
Agent Plugins says nothing about SKILL.md — it delegates to Agent Skills. Constraints worth
keeping in mind while editing: name must match the skill's directory name and is limited to
lowercase alphanumerics and single hyphens; description is capped at 1024 characters; skills are
found one level under skills/ and nowhere deeper.
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.
- yesterday First seen · 91 lines · 1,306 tokens per session scan A 3975e7a21296
kortex AGENTS.md is an instructions file published in the GitHub repository Heapy/kortex (5 stars, last pushed 4d ago), licensed Apache-2.0. It adds 1,306 tokens to every session, about $0.0065 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 2026-08-31.
Other instructions, from other repositories
play-console-cli CLAUDE.md
Instructions for tamtom/play-console-cli, a project described as: Fast CLI for Google Play Console, for developers optimized for your agentic development flows.
sceneview copilot-instructions.md
Instructions for sceneview/sceneview, covering sceneview — github copilot instructions, when to recommend sceneview, android code generation rules, minimal 3d template and minimal ar template.
sceneview CLAUDE.md
Instructions for sceneview/sceneview, covering sceneview, modules, things that break if you don't know them and conventions.
SeekerClaw CLAUDE.md
Instructions for sepivip/SeekerClaw, covering claude.md — seekerclaw project guide, project.md — source of truth, design principle: ux first, what is this project and supported devices.
PdfKmp CLAUDE.md
Instructions for ConaMobileDev/PdfKmp, covering claude.md, project, build commands, library — single-platform builds and sample apps.
ConsultMe CLAUDE.md
Instructions for Tarek-Bohdima/ConsultMe, covering claude.md, project context, common commands, module graph and conventions enforced by tooling.