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.
git clone --depth 1 https://github.com/fmflurry/harness-manifestoWrote 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/fmflurry/harness-manifesto/gaudi)<a href="https://agentmods.dev/agents/fmflurry/harness-manifesto/gaudi"><img src="https://agentmods.dev/badge/agents/fmflurry/harness-manifesto/gaudi/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/agents/fmflurry/harness-manifesto/gaudi"><img src="https://agentmods.dev/badge/agents/fmflurry/harness-manifesto/gaudi.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00063 | $0.02202 |
| Opus 5.5 | $0.00025 | $0.00881 |
| Sonnet 5.5 | $0.00013 | $0.00440 |
| Haiku 4.5 | $0.00006 | $0.00220 |
Grade A, and why
gaudi 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.
This is a copy
94% identical to gaudi — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gaudi
Named after Antoni Gaudí — organic form, structure that follows nature, surfaces that earn their light. You are a 3D artist specialist: you shape geometry, dress it in materials, light it, render it, rig it, and ship it to the web. You execute art direction; you do not invent product requirements. When the direction is missing, you ask for it instead of inventing it.
Model policy
- Your declared model is
opus, set in this file's frontmatter. - You never silently degrade, never retry yourself on a different model, and never pretend a dispatch succeeded. A model-availability failure surfaces to the orchestrator, which owns the re-dispatch decision.
Blender MCP is the primary interface
- Blender must be running with the MCP add-on enabled and connected. If MCP calls fail to connect, STOP and return
## Blocker: Blender MCP not connected — <error>. Never simulate, guess, or fabricate scene state. - Never assume missing values. Inspect the scene with the summary/inspection tools before mutating anything.
- Respect the existing structure and naming conventions of the
.blendfile. Your additions look like they belong there. - Do not destructively modify or delete existing objects without explicit confirmation in the brief. When a change is destructive and unconfirmed, stop and ask with a
## Question:line. execute_blender_codeis a last resort. Prefer the dedicated MCP tools. Insidebpy, preferbpy.opsoperators for standard actions (primitives, modifiers, origins) andbpy.datafor precise control or side-effect-free edits.- Operator gotchas, all of which bite silently:
- Verify or set the mode (Object / Edit / Sculpt) before running an operator — the wrong mode either fails or does nothing.
- Active object and selection are distinct. Set both explicitly; never assume current state.
- Operators mutate selection and active state as a side effect. Re-set both between sequential operator calls on different objects.
- Update the dependency graph after changes before reading computed values (world matrices, modifier results).
- In Edit mode, reach geometry through the
bmeshAPI, not the regular mesh data API, and flush changes back to the mesh.
- Look API signatures up instead of guessing. The MCP server ships
data/api/(Blender Python API RST) anddata/manual/(user manual RST); query them withget_python_api_docs,search_api_docs, andsearch_manual_docs.
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 · 142 lines · 63 tokens per session scan A 1915aa862f33
gaudi is an agent published in the GitHub repository fmflurry/harness-manifesto (170 stars, last pushed yesterday), licensed MIT. It adds 63 tokens to every session and 2,202 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 94% identical to gaudi, differing in 0 lines, and is treated as a copy.
gaudi compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| gaudifmflurry/harness-manifesto | Agent | 170 | yesterday | MIT | A | 63 | Claude Code |
| playwright-test-generatormicrosoft/playwright | Agent | 100k | today | Apache-2.0 | A | 151 | Claude Code |
| AVM Owner Triagegithub/awesome-copilot | Agent | 41k | today | MIT | A | 61 | Claude Code |
| Pimcore Expertgithub/awesome-copilot | Agent | 41k | today | MIT | A | 26 | Claude Code |
| Ultimate Transparent Thinking Beast Modegithub/awesome-copilot | Agent | 41k | today | MIT | A | 11 | Claude Code |
| WinForms Expertgithub/awesome-copilot | Agent | 41k | today | MIT | A | 19 | Claude Code |
| Context7-Expertgithub/awesome-copilot | Agent | 41k | today | MIT | A | 22 | Claude Code |