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 richfrem/agent-plugins-skills --skill local-llm-bridgegit clone --depth 1 https://github.com/richfrem/agent-plugins-skillsWrote 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/richfrem/agent-plugins-skills/local-llm-bridge)<a href="https://agentmods.dev/skills/richfrem/agent-plugins-skills/local-llm-bridge"><img src="https://agentmods.dev/badge/skills/richfrem/agent-plugins-skills/local-llm-bridge/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/richfrem/agent-plugins-skills/local-llm-bridge"><img src="https://agentmods.dev/badge/skills/richfrem/agent-plugins-skills/local-llm-bridge.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 | $0.00085 | $0.01168 |
| Opus 5 | $0.00043 | $0.00584 |
| Sonnet 5 | $0.00017 | $0.00234 |
| Haiku 4.5 | $0.00009 | $0.00117 |
Grade A, and why
local-llm-bridge scanned grade A with 1 finding 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 5d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
> **Requires llama-server running on port 8089.** Check: `curl http://localhost:8089/health` How it starts
The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Identity: The Local Gemma Sub-Agent Dispatcher
Dispatches bounded tasks directly to the optimized local Gemma 4 12B server at http://localhost:8089/v1/chat/completions. No routing proxy involved. Uses the run_agent.py task router with cli=llama.
[!IMPORTANT] Requires llama-server running on port 8089. Check:
curl http://localhost:8089/healthStart:./run_server.shin the local-llm-bench workspace. Thinking is disabled server-side (--reasoning off) — no special flags needed.
Why This Is Fast
The routing proxy (Mode A) carries ~29K tokens of Claude Code system prompt — at ~30 tok/s prefill that costs 60+ seconds per context boundary crossing.
This skill (Mode B) sends only the task prompt — typically 50–500 tokens. At 7+ tok/s generation on M1 Metal with a small context:
| Output length | Typical response time |
|---|---|
| 50 tokens | ~7s |
| 100 tokens | ~14s |
| 200 tokens | ~28s |
Default max_tokens=120 keeps responses terse. Override via code if needed.
Orchestration Pattern: run_agent.py
python ./scripts/run_agent.py \
<PERSONA_FILE> <INPUT_FILE> <OUTPUT_FILE> "<INSTRUCTION>" \
--cli llama --max-tokens 120
Example — code review
python ./scripts/run_agent.py \
agents/refactor-expert.md \
target.py \
review.md \
"List the top 3 issues in this code. Be terse." \
--cli llama
Example — summarize a diff (longer output)
python ./scripts/run_agent.py \
/dev/null \
changes.diff \
summary.md \
"Summarize this diff in 2 sentences. Focus on risk." \
--cli llama --max-tokens 200
Example — instruction only (no input file)
python ./scripts/run_agent.py \
/dev/null /dev/null \
answer.md \
"What is the capital of France? One word." \
--cli llama --max-tokens 10
Prompt Budget Guidelines
Keep prompts lean — this is the primary performance lever:
- Persona: 100–300 tokens (enough to set role and tone)
- Source file: keep under 2,000 tokens where possible; trim to the relevant section
- Instruction: 1–3 sentences; specific and bounded
- Expected output: terse — list form, not prose paragraphs
What ships with it
10 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.
- evals/evals.json 614 B
- evals/results.tsv 34 B
- references/acceptance-criteria.md 42 B
- references/cheapest_models.json 40 B
- references/cheapest_models.md 38 B
- requirements.txt 0 B
- scripts/kv_cache_orchestrator.py 41 B runs code
- scripts/run_agent.py 29 B runs code
- scripts/run_server.py 30 B runs code
- scripts/test_run_agent.py 34 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.
- 5d ago First seen · 149 lines · 85 tokens per session scan A 759e2193748d
local-llm-bridge is a skill published in the GitHub repository richfrem/agent-plugins-skills (6 stars, last pushed today), licensed MIT. It adds 85 tokens to every session and 1,168 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
swarm-scripts
Bulk, repeat, fan-out, or data-heavy work: write and run swarm scripts (inline script-run, named script-upsert, durable launch-script-run). Covers the script-vs-tool rubric, the authoring contract (args first, ctx second), the seed catalog, connections and secrets, dbquery, and exposing a script as an API.
semantic-creator
A workflow for turning APIs, command-line tools, or database tables into a governed semantic layer—a shared description of data and its meaning. It uses Kimball-style dimensional modeling and can produce OKF or YAML files after review.
prompt-decompiler
This skill should be used when the user says "decompile this prompt", "reverse engineer this prompt", "analyze this prompt", "break down this prompt", "what does this prompt do", "adapt this prompt for", "simplify this prompt", "split this prompt", "migrate this prompt", "why does this prompt work", "explain this…
model-cli-gemini
Use when invoking Google's Gemini directly through the shared Antigravity, Gemini CLI, or agent fallback chain.
custom-model-training
Trains a custom Ideogram model on a folder of reference images, then generates a new image with that trained model to prove it works — a real end-to-end pipeline (createdataset, uploaddatasetassets, trainmodel, poll getmodel, generateimage with custommodeluri), not just an explanation of how training works. Use…
ideogram-prompt
Prompting technique guide for Ideogram image generation via the connected Ideogram MCP (generateimage, describeimage, remiximage, editimage). Use whenever the user wants to generate, remix, or edit an image with Ideogram — including moodboards, style-matching a reference image, extracting a visual "recipe" from…