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 gaabrielrd/local-model-workers-mcp --skill offload-to-local-modelsgit clone --depth 1 https://github.com/gaabrielrd/local-model-workers-mcpWrote 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/gaabrielrd/local-model-workers-mcp/offload-to-local-models)<a href="https://agentmods.dev/skills/gaabrielrd/local-model-workers-mcp/offload-to-local-models"><img src="https://agentmods.dev/badge/skills/gaabrielrd/local-model-workers-mcp/offload-to-local-models/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/gaabrielrd/local-model-workers-mcp/offload-to-local-models"><img src="https://agentmods.dev/badge/skills/gaabrielrd/local-model-workers-mcp/offload-to-local-models.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 | $0.00082 | $0.00729 |
| Opus 5.5 | $0.00033 | $0.00292 |
| Sonnet 5.5 | $0.00016 | $0.00146 |
| Haiku 4.5 | $0.00008 | $0.00073 |
Grade A, and why
offload-to-local-models 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 2mo 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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Offload repository work to local model tools
The local-model-workers MCP server delegates expensive repository work to a
model running on the user's own machine or LAN. Every tool is read-only against
the repository: patches come back as validated, unapplied unified diffs that you
decide whether to apply.
Using these tools instead of scanning files yourself keeps the conversation short and keeps repository content on the user's network.
Rules
- Do not echo large tool results verbatim into the conversation. Reference paths and summarize findings instead.
- Prefer a targeted
query_code_graph,search_semantic, orsummarize_modulecall over a broadexplore_repositorywhen a focused lookup suffices. - Every tool takes an absolute
repository_root. Pass the project root, not a subdirectory, unless the user asked to scope the work. - Patches are proposals. Show the diff and let the user decide; never apply one silently.
Which tool to reach for
| Situation | Tool |
|---|---|
| Understand an unfamiliar area, goal stated in prose | explore_repository |
| Find code by intent ("where do we retry failed uploads?") | search_semantic |
| Callers, dependencies, exports, or definition of a symbol | query_code_graph |
| Structured summary of one file or directory | summarize_module |
| Propose unit tests as an unapplied diff | propose_tests |
| Generate tests and iterate until they pass in a sandbox copy | auto_validate_tests |
Docstrings, JSDoc, or a docs/<slug>.md guide |
generate_docs_patch |
| Explain what a commit range changed and its architectural impact | analyze_diff |
| Repair reported linter violations | fix_lint_violations (pass the linter output) |
| Repair compiler or type-checker errors | fix_type_errors (pass the checker output) |
| Provider reachability, model routing, configuration | check_health, get_config |
| Tokens and queries saved by offloading | get_offload_stats |
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.
- 2mo ago First seen · 60 lines · 82 tokens per session scan A 8ad45ff69c46
offload-to-local-models is a skill published in the GitHub repository gaabrielrd/local-model-workers-mcp (1 stars, last pushed 3mo ago), licensed MIT. It adds 82 tokens to every session and 729 once invoked, about $0.0003 per session on Opus 5.5. 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.
offload-to-local-models compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| offload-to-local-modelsgaabrielrd/local-model-workers-mcp | Skill | 1 | 3mo ago | MIT | A | 82 | Claude Code |
| imagegenopenai/codex | Skill | 134k | today | Apache-2.0 | A | 113 | Codex |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | yesterday | MIT | A | 200 | Claude Code, Codex |
| playwright-climicrosoft/playwright | Skill | 100k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 394k | today | MIT | A | 36 | Claude Code |
| systematic-debuggingobra/superpowers | Skill | 314k | 4d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |