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 Rockielab/rockie-claude --skill inference-engineergit clone --depth 1 https://github.com/Rockielab/rockie-claudeWrote 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/rockielab/rockie-claude/inference-engineer)<a href="https://agentmods.dev/skills/rockielab/rockie-claude/inference-engineer"><img src="https://agentmods.dev/badge/skills/rockielab/rockie-claude/inference-engineer/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/rockielab/rockie-claude/inference-engineer"><img src="https://agentmods.dev/badge/skills/rockielab/rockie-claude/inference-engineer.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 | $0.00159 | $0.07056 |
| Opus 5 | $0.00079 | $0.03528 |
| Sonnet 5 | $0.00032 | $0.01411 |
| Haiku 4.5 | $0.00016 | $0.00706 |
Grade A, and why
inference-engineer 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- inference-engineer — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 346 lines — stays where its author put it; the contents beside it link to each section on GitHub.
inference-engineer
The packaging skill that turns "here is a model" into "here is an API your agent can call." Companion to /autoresearch (Rockie's R&D half): autoresearch produces models and findings, inference-engineer operates them.
When to invoke
- Explicit:
/inference-engineer <model-url-or-HF-repo>(URL optional; agent will ask for it). - Intent-triggered: any user message expressing one of —
- "I want to host / serve / deploy / productize / run inference on / get an API for / make available "
- "I trained / fine-tuned a model — what do I do with it?"
- "How do I serve this for my agent / app / customers / a batch of N requests?"
- "Reproduce paper / blog / GitHub repo X on a small GPU"
- Cascade entry from sub-skills: the eval / kernel / gateway sub-skills can call back here to re-provision when a user asks to swap hardware or change serving config.
The two researcher motivations this skill serves
- "I found a model — productize it." A researcher has a HF repo / GitHub repo / paper-with-code link. They want to host it themselves (cost, privacy, latency, customization). They need: right hardware, right serving stack, an API, MCP exposure to their chat agent, quality + cost numbers.
- "I trained a model — now what?" Autoresearch produced a checkpoint. The researcher wants to share it, evaluate it head-to-head against the baseline, or wire it into a downstream agent. They need the same productization path but starting from local weights instead of a public repo.
Both routes converge on the same 8-step orchestration below.
If this skill needs Rockie-managed training, eval, synthetic-data, or
other GPU job execution outside POST /api/inference/loads, route that
work through /experiment and its embedded budget-term-sheet approval gate.
Do not introduce raw /api/jobs/submit calls here.
Read this before doing anything
The skill reasons from a checked-in research corpus, not from re-Googling. Before step 1 of any run, read all of:
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- prompts/cost-confirm.md 3.0 KB
- prompts/intake-clarify.md 4.8 KB
- research/api-gateways.md 5.8 KB
- research/evaluation.md 6.1 KB
- research/hardware-workload.md 5.8 KB
- research/kernel-engineering.md 7.3 KB
- research/serving-stacks.md 6.0 KB
- runtime/monitor_contract.py 3.6 KB runs code
- test_skill_auth_contract.py 790 B runs code
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 · 346 lines · 159 tokens per session scan A 62ae8834a2a9
inference-engineer is a skill published in the GitHub repository Rockielab/rockie-claude (21 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 159 tokens to every session and 7,056 once invoked, about $0.0008 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
jsonld-knowledge-graph
Design and ship a companion JSON-LD knowledge graph (graph.jsonld) next to llms.txt for projects with stable concept-level structure. Encodes domain entities and relationships as schema.org triples for LLM citation. Use when project has matrix / hierarchy / phase-binding structure that prose alone leaves implicit, AND…
llm-as-judge
Design pattern for LLM-as-judge evaluators — binary checks as evidence, one named holistic verdict, no score aggregation. Use when designing or reviewing any LLM-based quality gate, evaluator, judge prompt, or verdict schema; when a judge's rubric scores fluctuate between runs; when you catch yourself asking an LLM…
prompt-perturb
An idea-generation tool that fetches creative prompts from outside sources after removing project-specific context from the search.
constitutional-ai
Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with self-critique/revision, then RLAIF (RL from AI Feedback). Use for safety alignment, reducing harmful outputs without human labels. Powers Claude's safety system.
auto-run
Autonomous personalized research loop. Use when the user wants to research a topic autonomously, run a research loop, start adaptive research, or use presets like technique-scout or cross-domain. Triggers on: 'auto run', 'research loop', 'autonomous research', 'run research', 'start research', 'adaptive research'.
ai-evaluation
Systematic evaluation (evals) for LLM and AI products. Design test cases, measure accuracy/quality, track regressions, benchmark models, and build continuous evaluation pipelines. Distinct from traditional software testing with probabilistic outputs.