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 Robotti-io/copilot-security-instructions --skill genai-acceptance-reviewgit clone --depth 1 https://github.com/Robotti-io/copilot-security-instructionsWrote 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/robotti-io/copilot-security-instructions/genai-acceptance-review)<a href="https://agentmods.dev/skills/robotti-io/copilot-security-instructions/genai-acceptance-review"><img src="https://agentmods.dev/badge/skills/robotti-io/copilot-security-instructions/genai-acceptance-review/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/robotti-io/copilot-security-instructions/genai-acceptance-review"><img src="https://agentmods.dev/badge/skills/robotti-io/copilot-security-instructions/genai-acceptance-review.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 | $0.00028 | $0.00549 |
| Opus 5 | $0.00014 | $0.00275 |
| Sonnet 5 | $0.00006 | $0.00110 |
| Haiku 4.5 | $0.00003 | $0.00055 |
Grade A, and why
genai-acceptance-review 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 12d 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GenAI Acceptance Review
When to use
Use this skill when a system consumes LLM output to make decisions or perform actions.
Inputs to collect (if available)
- What the model output is used for (advisory vs actionable)
- Tools/capabilities available to the system (file writes, network calls, deploys)
- Data entering prompts (PII/secrets? retrieved content sources?)
- Approval model (human-in-the-loop? step-up auth?)
Threats to consider
- Prompt injection (content causes the model to ignore instructions)
- Over-trust / tool misuse (model output drives privileged actions)
- Data leakage (secrets/PII included in prompts or outputs)
- Indirect injection via retrieved content (RAG, web pages, PDFs)
Step-by-step process
- Map the AI boundary
- Where prompts are built, where tools are called, what data enters/leaves.
- Classify outputs
- Advisory: suggestions for humans
- Actionable: used by code to execute, write files, call APIs, change permissions
- Apply controls by class
- Advisory: disclaimers, human review, logging with redaction
- Actionable: strict schema validation, allow-lists, capability gating, step-up approvals
- Prompt & retrieval hardening
- Separate system instructions from untrusted content
- Use structured output (JSON schema) and reject invalid outputs
- Limit context sources; sanitize retrieved content where possible
- Add misuse tests
- Include injection strings and verify they don’t trigger privileged actions
- Document safe usage
- Clear rules for what the model may decide vs what code must enforce
Output
- Boundary diagram (textual is fine)
- Control recommendations (prevent/detect/respond)
- Test cases for injection and over-trust scenarios
Repo integration (optional)
Related prompt:
check-for-unvalidated-genai-acceptances.prompt.md
Output format
- Boundary map: where untrusted content enters, where model output leaves
- Threats: top 5 with likelihood/impact
- Controls: prevent/detect/respond mapped to advisory vs actionable use
- Validation: misuse/prompt-injection test scenarios
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.
- 12d ago First seen · 67 lines · 28 tokens per session scan A 50bc49ec1f8c
genai-acceptance-review is a skill published in the GitHub repository Robotti-io/copilot-security-instructions (42 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 28 tokens to every session and 549 once invoked, about $0.0001 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
create-simple-prompt
This skill should be used when the user asks to "create a new prompt sample", "add a new prompt sample", "scaffold a new prompt sample", "create a prompt contribution", "add a prompt", or needs to create a new prompt sample with proper folder structure, README, and sample.json metadata. Do NOT use this skill for agent…
prompt-context-engineer
Transform any rough prompt, request, or idea into a well-structured, context-engineered prompt using Andrej Karpathy's context engineering principles. Use this skill whenever the user asks to improve, rewrite, restructure, optimize, or "engineer" a prompt; mentions prompt engineering or context engineering; pastes a…
vendor-prompting
ANALYSIS SKILL — Audit-grade reference for Anthropic Claude and OpenAI GPT-5.6 prompting best practices. WHEN: "claude prompting", "gpt-5.6 prompting", "audit agent", "review prompt", "vendor best practices", "anthropic best practices", "openai prompting". DO NOT USE FOR: routine prompt edits where rules are already…
deploy-model
Unified Azure OpenAI model deployment skill with intelligent intent-based routing. Handles quick preset deployments, fully customized deployments (version/SKU/capacity/RAI policy), and capacity discovery across regions and projects. USE FOR: deploy model, deploy gpt, create deployment, model deployment, deploy openai…
azure-kusto
ANALYSIS SKILL — Query and analyze data in Azure Data Explorer (Kusto/ADX) using KQL. WHEN: "KQL queries", "Kusto database queries", "Azure Data Explorer", "ADX clusters", "time series data", "IoT telemetry", "anomaly detection". DO NOT USE FOR: App Insights / Log Analytics troubleshooting (azure-diagnostics), cost…
image-prompt-builder-nl
Craft high-quality natural-language image prompts for any modern text-to-image or image-edit model that accepts flowing English. Trigger when the user wants help writing, rewriting, improving, or translating an English natural-language image prompt — including "write me an image prompt", "improve this image prompt"…