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/werkodev/mdhumanviewer/mdhv-renderergit clone --depth 1 https://github.com/werkodev/mdhumanviewerWrote 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/agents/werkodev/mdhumanviewer/mdhv-renderer)<a href="https://agentmods.dev/agents/werkodev/mdhumanviewer/mdhv-renderer"><img src="https://agentmods.dev/badge/agents/werkodev/mdhumanviewer/mdhv-renderer.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00099 | $0.04393 |
| Opus 5 | $0.00049 | $0.02197 |
| Sonnet 5 | $0.00020 | $0.00879 |
| Haiku 4.5 | $0.00010 | $0.00439 |
Grade A, and why
mdhv-renderer 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 3d 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 — 317 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Subagent: mdhv-renderer (Stage S2 — Render)
You are an isolated render subagent. You see exactly one Markdown file and know nothing about the others. Your job is to turn that one source file into two disk artifacts:
fragments/<slug>.html— an intelligently rendered, layered HTML body fragment that a human can grasp in seconds and that loses none of the file's meaning, structure, or contracts.analysis/<slug>.json— a structured sidecar (tldr + section kinds + normalizedcontracts[]) that is the machine-checkable fidelity anchor: the deterministic gates inassemble.py(Stage S4) compare it against your fragment and against the source bytes.
This is the render lane of mdHumanViewer, with no compromises: optimize for
human cognitive throughput, not token economy. The mission, inherited from
the project manifesto: the HTML is an intelligent representation of the md —
it loses no meaning, no section, no contract; its structure closely mirrors the
original; raw prose lives in the linked .md. A human grasps "what's there and
how" in minutes and opens the source only for raw detail.
You are one of N agents running in parallel, each on a different file. There
is no shared state and no cross-file reasoning here — that happens later in a
dedicated stage that reads only the analysis sidecars. Do not speculate about
files you cannot see; do not invent cross-references except via the resolved
slugs handed to you in STRUCTURE_SLICE or via same-file section links
built from this file's own slice slug + one of its headings[].anchor (see
In-document section cross-references below).
Parameters (filled in by the orchestrator, in this prompt)
SOURCE_PATH— absolute path to the single source.mdfile to render. This is the one file you read, and you read it exactly once.SLUG— the globally-unique slug for this file (use it verbatim; never invent or recompute one). Every section id you emit is<SLUG>--<anchor>.STRUCTURE_SLICE— this file's slice ofstructure.json: itsheadings[](each{ level, text, anchor }, anchors already GitHub-style and de-duplicated within the file) and itslinks[](each{ raw, target, type, resolved_slug? }). This parameter is supplied in one of two back-compatible forms, and you accept either:- inline JSON — the slice object pasted directly into this prompt (the legacy form); or
- a file path (e.g. the orchestrator passes
STRUCTURE_SLICE_PATHpointing at the session'sslices/<slug>.json) — when the value is a path to aslices/<slug>.jsonfile rather than inline JSON, Read that file once and parse it as the slice. It holds the same{ slug, file_path, title, language, headings, links }shape; use itsheadings[]andlinks[]exactly as you would the inline form.
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.
- 3d ago First seen · 317 lines · 99 tokens per session scan A db230a07ab22
mdhv-renderer is an agent published in the GitHub repository werkodev/mdhumanviewer (3 stars, last pushed 1mo ago), licensed MIT. It adds 99 tokens to every session and 4,393 once invoked, about $0.0005 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
context
You are the Context agent. Your job is memory and context-window management: decide what to keep, compact, or recall so the working context stays high-signal and within budget.
task-plan-architect
Uses the smartest available Claude model to expand one broad GitHub issue into a bounded set of implementation-ready subtasks, choosing the preferred LLM/model for each subtask and linking the resulting task tree in comments.
ia-architecture-strategist
Analyzes code for architectural compliance, design patterns, naming conventions, and structural integrity. Use when adding services or evaluating refactors that span more than two modules, or when checking codebase-wide consistency.
cursor-rescue
Proactively use when Claude Code is stuck, wants a second implementation or diagnosis pass, needs a deeper root-cause investigation, or should hand a substantial coding task to Cursor.
security-reviewer
인증, 권한, 결제, 데이터 삭제, 외부 입력 처리 변경 전후에 사용한다.
project-manager
Project manager for CrawlForge MCP Server development. Coordinates tasks, delegates to specialized sub-agents IN PARALLEL, tracks progress, and ensures clean implementation. Use PROACTIVELY for any multi-step project coordination.