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 agents/dsswift/ion/definition-formatgit clone --depth 1 https://github.com/dsswift/ionWhat 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 | $0.00009 | $0.00653 |
| Opus 5 | $0.00005 | $0.00327 |
| Sonnet 5 | $0.00002 | $0.00131 |
| Haiku 4.5 | $0.00001 | $0.00065 |
Grade A, and why
definition-format 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 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.
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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Definition Format
Agent definitions are markdown files (.md) with YAML frontmatter. The frontmatter declares agent metadata; the markdown body becomes the agent's system prompt.
File structure
---
name: code-reviewer
parent: main
description: Reviews code changes for correctness and style
model: claude-haiku-4-5-20251001
tools: [Read, Grep, Glob]
---
You are a code review specialist. Review the provided code changes
for correctness, style consistency, and potential bugs. Focus on
logic errors and missing edge cases.
Frontmatter fields
| Field | Required | Description |
|---|---|---|
name |
No | Agent name. Defaults to the filename stem (e.g., code-reviewer.md becomes code-reviewer). |
parent |
No | Parent agent name. Empty means root agent. See hierarchies. |
description |
No | One-line description. Used in agent listings and tool descriptions. |
model |
No | Model override for this agent. When omitted, inherits the session's model. See multi-model. |
tools |
No | Tool allowlist. Accepts inline array [Read, Grep, Glob] or comma-separated values. When omitted, the agent has access to all tools. |
Any frontmatter key not listed above is stored in the agent's Meta map, accessible to extensions but not used by the engine directly.
System prompt body
Everything after the closing --- fence becomes the agent's system prompt. The engine passes this as the system message when the agent runs.
The system prompt supports any markdown formatting. It is sent to the LLM as-is, with no template interpolation.
Frontmatter parsing rules
- The file must start with
---(leading whitespace is trimmed). - The frontmatter block ends at the next
\n---boundary. - Lines starting with
#inside the frontmatter are treated as comments and ignored. - The
toolsfield accepts both[a, b, c]and barea, b, csyntax.
Name resolution
When the name field is omitted, the engine derives the name from the file path:
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.
- 2d ago First seen · 76 lines · 9 tokens per session scan A 4a7ec466bd96
definition-format is an agent published in the GitHub repository dsswift/ion (4 stars, last pushed 2d ago), licensed MIT. It adds 9 tokens to every session and 653 once invoked, about $0.0000 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 agents, from other repositories
document_agent
Minimal example for creating and using AI Agents with aworld-cli.
building-agents
Create intelligent agents that connect to MCP servers, discover tools automatically, and execute tasks with full observability.
reasoning-patterns
Every Promptise agent is powered by a Reasoning Graph. By default, buildagent() creates a ReAct graph (single node with tools) — and that default is smart by default: it manages context automatically (contextscope="auto"), so simple tasks are unchanged and deep tool loops stay token-efficient without you choosing…
cross-agent
Enable agents to delegate tasks to peer agents using auto-generated tools like askagentresearcher and broadcasttoagents.
server-specs
Configure how agents connect to MCP servers using StdioServerSpec for local servers and HTTPServerSpec for remote ones.
superagent-files
Define agents declaratively using .superagent YAML files -- configure model, servers, memory, sandbox, and cross-agent references in a single file.