Horse is a lightweight web framework for Delphi and Lazarus programs, providing tools for building HTTP servers and APIs. It is for developers who need routing, request handling, middleware, streaming, WebSockets, and related server features in those languages. Its catalogue skills support working with the framework.
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 skills add HashLoad/horse --skill horse-writing-middlewaregit clone --depth 1 https://github.com/HashLoad/horseWrote 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.
[](https://agentmods.dev/skills/hashload/horse/horse-writing-middleware)<a href="https://agentmods.dev/skills/hashload/horse/horse-writing-middleware"><img src="https://agentmods.dev/badge/skills/hashload/horse/horse-writing-middleware/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.
<a href="https://agentmods.dev/skills/hashload/horse/horse-writing-middleware"><img src="https://agentmods.dev/badge/skills/hashload/horse/horse-writing-middleware.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00025 | $0.00337 |
| Opus 5 | $0.00013 | $0.00169 |
| Sonnet 5 | $0.00005 | $0.00067 |
| Haiku 4.5 | $0.00003 | $0.00034 |
Grade A, and why
horse-writing-middleware 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 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.
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.
What it actually says
Writing Custom Horse Middlewares
Middleware Signature
A custom middleware is a procedure that receives THorseRequest, THorseResponse, and a callback procedure Next.
procedure MyCustomMiddleware(Req: THorseRequest; Res: THorseResponse; Next: TProc);
begin
// 1. Logic BEFORE the request reaches the endpoint handler
Req.Headers.AddOrSetValue('X-Processed-By', 'MyMiddleware');
// 2. Call Next to delegate execution to the subsequent middleware or route handler
Next;
// 3. Logic AFTER the route handler completes its execution
Res.Headers.AddOrSetValue('X-Execution-Finished', 'True');
end;
Terminating Execution Early
If a validation fails (e.g., authentication is missing), do NOT call Next. Set the status and send the response immediately. This halts the pipeline execution.
procedure AuthMiddleware(Req: THorseRequest; Res: THorseResponse; Next: TProc);
var
LToken: string;
begin
LToken := Req.Headers.Items['Authorization'];
if LToken.IsEmpty then
begin
// Halt execution by sending response and omitting the Next call
Res.Send('Unauthorized').Status(THTTPStatus.Unauthorized);
Exit;
end;
// Token is valid, proceed with pipeline
Next;
end;
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.
- 10d ago First seen · 48 lines · 25 tokens per session scan A 37bb1b7bc505
horse-writing-middleware is a skill published in the GitHub repository HashLoad/horse (1,373 stars, last pushed today), licensed MIT. It adds 25 tokens to every session and 337 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
api-design-patterns
Comprehensive API design patterns covering REST, GraphQL, gRPC, versioning, authentication, and modern API best practices.
api-design-patterns
Comprehensive API design patterns covering REST, GraphQL, gRPC, versioning, authentication, and modern API best practices.
api-design-patterns
Comprehensive API design patterns covering REST, GraphQL, gRPC, versioning, authentication, and modern API best practices.
hunt-wordpress
Use when an authorized target exposes WordPress core, plugin, theme, REST, or XML-RPC behavior.
mars-development
Develop REST APIs with MARS-Curiosity (Delphi REST library) - writing resources, REST attributes, parameter binding, JWT authentication and roles, FireDAC dataset publishing, server-sent events (SSE), WebStencils HTML templating, client components, configuration, serialization. Use this skill whenever the user is…
a0-development
Development guide for extending Agent Zero from current source and DOX. Use for framework architecture, tools, extensions, API/WebUI handlers, agent profiles, prompts, skills, projects, runtime boundaries, and contribution workflow. Load the focused reference files before giving implementation guidance.