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 CarbeneAI/Forge --skill privategit clone --depth 1 https://github.com/CarbeneAI/ForgeWrote 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/carbeneai/forge/private)<a href="https://agentmods.dev/skills/carbeneai/forge/private"><img src="https://agentmods.dev/badge/skills/carbeneai/forge/private/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/carbeneai/forge/private"><img src="https://agentmods.dev/badge/skills/carbeneai/forge/private.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 | $0.00070 | $0.00763 |
| Opus 5 | $0.00035 | $0.00381 |
| Sonnet 5 | $0.00014 | $0.00153 |
| Haiku 4.5 | $0.00007 | $0.00076 |
Grade B, and why
Private scanned grade B with 2 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 6d 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.
Sends data to an external URLmediumData exfiltration
A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.
curl http://localhost:11434/api/generate -d '{"model":"gemma4:31b","prompt":"hello","stream":false}' Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl http://localhost:11434/api/tags How it starts
The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Private - Local AI Privacy Router
Route queries to Gemma4 31B on your local Ollama server. Data never leaves your local machine.
How It Works
When invoked, this skill:
- Takes your prompt
- Sends it to Gemma4 31B via Ollama on
localhost:11434 - Returns the response inline — Claude orchestrates, but never sees your prompt content
Usage
/private What are the financial projections for client XYZ?
/private Analyze this medical report: [paste]
/private Review these employee performance notes
Workflow
When the user invokes /private:
- Extract the user's prompt (everything after
/private) - Run the query tool:
bun ${PAI_DIR}/skills/Private/tools/private-query.ts "<prompt>"
- Return the Ollama response to the user verbatim — do NOT summarize or rephrase it
Available Models
| Model | Size | Best For |
|---|---|---|
gemma4:31b (default) |
31.3B | General intelligence, reasoning, analysis |
dolphin3:8b-llama3.1-fp16 |
8.0B | Fast responses, uncensored |
dolphin3:8b |
8.0B | Fast responses, lower VRAM |
llama3.1:latest |
8.0B | General purpose |
Options
/private <prompt> # Default: gemma4:31b
/private --model dolphin3:8b <prompt> # Use specific model
/private --system "You are a CFO" <prompt> # Custom system prompt
/private --temperature 0.3 <prompt> # Lower temp for precision
Privacy Guarantees
- Data stays on your local machine — never sent to Anthropic, OpenAI, or any cloud
- No logging by third parties
- No API keys transmitted
- Suitable for HIPAA/PCI-sensitive data
- Ollama runs on localhost:11434 (not exposed externally by default)
When to Use
| Scenario | Use /private? |
|---|---|
| Client financial data | Yes |
| Medical/health information | Yes |
| Employee reviews or HR data | Yes |
| Proprietary business strategy | Yes |
| Code with trade secrets | Yes |
| General coding questions | No — use Claude |
| Web research needed | No — use Claude/Research |
| Latest information needed | No — Ollama has training cutoff |
What ships with it
1 file 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.
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.
- 6d ago First seen · 96 lines · 70 tokens per session scan B 186431b9c116
Private is a skill published in the GitHub repository CarbeneAI/Forge (9 stars, last pushed 1mo ago), licensed MIT. It adds 70 tokens to every session and 763 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 2 findings (sends data to an external url, 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
update-llm-models
Autonomously research and update the default lightweight LLM models in src/core/llm.ts using web search to ensure the most current versions are used.
llm-redteam-overview
LLM red team category — full AATMF v3 tactic coverage (T01–T15). Routing skill: read this first to identify which tactic applies, then load the matching sub-skill. Maps to MITRE ATLAS where overlap exists.
implementing-llm-guardrails-for-security
Implements input and output validation guardrails for LLM-powered applications to prevent prompt injection, data leakage, toxic content generation, and hallucinated outputs. Builds a security validation pipeline using NVIDIA NeMo Guardrails Colang definitions, custom Python validators for PII detection and content…
prompt-engineer
Prompt engineering expert for chain-of-thought, few-shot learning, evaluation, and LLM optimization.
prompt-lab
LLM prompt engineering: analyzes failure modes, generates variants (direct, few-shot, CoT), designs rubrics, produces test suites. Triggers on: "prompt engineering", "generate prompt variants", "A/B test prompts", "optimize prompt", "improve this prompt". NOT for SKILL.md files, use skill-evaluator.
Prompt Engineering
Write better prompts for AI tools — structured, specific, and optimized for quality output.