ai-risk-management

Name: ai-risk-management
Author: briiirussell

320stars36forks

How to install ai-risk-management?

briiirussell/cybersecurity-skills/ai-risk-management

$npx -y skills add briiirussell/cybersecurity-skills --skill ai-risk-management

Installs into the current project.

Summary

Apply the NIST AI Risk Management Framework (AI RMF 1.0) and adjacent guidance to AI / ML systems — model lifecycle governance, fairness and bias evaluation, robustness, transparency, accountability, third-party model risk, monitoring for drift, and AI incident response. Broader

Files · 1

View on GitHub

SKILL.md

1	# AI Risk Management — Beyond Security, the Whole Model Lifecycle
2
3	`prompt-injection` covers the AI security slice — attackers manipulating LLM inputs. This skill covers everything else risk-related about deploying AI / ML systems: governance, fairness, robustness, transparency, monitoring, incident response specific to AI failures, third-party model risk, and compliance with the emerging AI regulatory landscape.
4
5	The framing is NIST AI RMF 1.0 (released 2023) — the most widely-adopted voluntary framework — plus the regulatory layer (EU AI Act, US executive orders, sector-specific guidance). Use this skill when you are deploying AI features beyond a chatbot wrapper, when a regulator asks "how do you govern your AI," or when something has gone wrong with an AI system in production.
6
7	Cross-references: `prompt-injection` for prompt-injection / LLM-specific security attacks; `threat-modeling` for design-time AI risk modeling; `incident-triage` and `breach-patterns` for AI-related incident response patterns; `csf-mapping` for the broader governance frame that AI RMF sits within.
8
9	## The NIST AI RMF — four functions
10
11	Just like the cybersecurity framework, the AI RMF organizes the work into functions. Same shape, different content.
12
13	\| Function \| What it covers \|
14	\|---\|---\|
15	\| Govern (GOV) \| Policy, accountability, roles, risk appetite, AI principles, board oversight, governance structures \|
16	\| Map (MAP) \| Context — what is the AI system, what does it do, who is impacted, what could go wrong, what are the legal / ethical constraints \|
17	\| Measure (MEAS) \| Evaluate the system — fairness, robustness, accuracy, explainability, privacy, security; quantitative + qualitative metrics \|
18	\| Manage (MAN) \| Treat the risks — mitigations, monitoring, incident response, decommissioning, ongoing review \|
19
20	The framework is voluntary but increasingly cited in contracts, RFPs, executive orders, and emerging regulations. Treat it as the lingua franca of AI risk.
21
22	## Workflow
23
24	### Step 1 — Inventory AI systems
25
26	Before assessment, build the inventory. Most organizations underestimate how much AI they actually deploy.
27
28	\| Category \| Examples \|
29	\|---\|---\|
30	\| First-party trained models \| Recommendation engines, fraud detection, churn prediction, internal ML pipelines \|
31	\| First-party LLM use \| Customer support chat, content generation, summarization, code generation, embeddings for search \|
32	\| Third-party AI features \| Stripe Radar (fraud), GitHub Copilot (code completion), Salesforce Einstein, Notion AI, Linear AI \|
33	\| Embedded AI in products you ship \| Suggested responses, smart defaults, AI sorting / ranking \|
34	\| AI in HR / hiring \| Resume screening, candidate matching, performance evaluation — high regulatory exposure \|
35	\| AI in customer-facing decisions \| Pricing, eligibility, content moderation, ad targeting — high regulatory exposure \|
36
37	For each, record: vendor (if any), training data source, deployment context, who it affects, the decision it informs, how decisions are reviewed.
38
39	### Step 2 — MAP: assess the context per system
40
41	For each AI system in the inventory, answer:
42
43	- Purpose — what is this system's stated goal? Does the actual deployment match?
44	- Stakeholders — who interacts with it, who is affected by its decisions, who is in a position to challenge those decisions?
45	- Legal / regulatory context — is this in scope for a specific law? (EU AI Act high-risk categories, US HUD fair-housing rules, EEOC for employment AI, FTC for unfair / deceptive practices, sector laws)
46	- Failure modes — what does "broken" look like? (Wrong answer, biased answer, hallucinated answer, slow answer, expensive answer, refused-to-answer-something-it-should, answered-something-it-should-not)
47	- Reversibility — when this system makes a wrong call, can the decision be undone? (Mortgage denial: hard to undo. Spam filter: easy)
48
49	### Step 3 — MEASURE: evaluate the system
50
51	The categories of evaluation, with the engineering hooks for each:
52
53	#### Accuracy / performance
54
55	- Test set evaluation — held-out data, not the training data
56	- Performance on slices of data, not just aggregate

briiirussellPublisher · 30 listings

Preview

briiirussell/cybersecurity-skills

# AI Risk Management — Beyond Security, the Whole Model Lifecycle

`prompt-injection` covers the AI security slice — attackers manipulating LLM inputs. This skill covers everything else risk-related about deploying AI / ML syst

The framing is NIST AI RMF 1.0 (released 2023) — the most widely-adopted voluntary framework — plus the regulatory layer (EU AI Act, US executive orders, sector

Cross-references: `prompt-injection` for prompt-injection / LLM-specific security attacks; `threat-modeling` for design-time AI risk modeling; `incident-triage`

Repobriiirussell/cybersecurity-skills

LicenseMIT

TypeSkills

CategorySecurity

UpdatedMay 2026

Repo stars320

First seenJul 22, 2026

People who liked this also viewed

6 picks

Type

ai-risk-management

320stars36forks

View on GitHub

How to install ai-risk-management?

briiirussell/cybersecurity-skills/ai-risk-management

$npx -y skills add briiirussell/cybersecurity-skills --skill ai-risk-management

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# AI Risk Management — Beyond Security, the Whole Model Lifecycle
2
3	`prompt-injection` covers the AI security slice — attackers manipulating LLM inputs. This skill covers everything else risk-related about deploying AI / ML systems: governance, fairness, robustness, transparency, monitoring, incident response specific to AI failures, third-party model risk, and compliance with the emerging AI regulatory landscape.
4
5	The framing is NIST AI RMF 1.0 (released 2023) — the most widely-adopted voluntary framework — plus the regulatory layer (EU AI Act, US executive orders, sector-specific guidance). Use this skill when you are deploying AI features beyond a chatbot wrapper, when a regulator asks "how do you govern your AI," or when something has gone wrong with an AI system in production.
6
7	Cross-references: `prompt-injection` for prompt-injection / LLM-specific security attacks; `threat-modeling` for design-time AI risk modeling; `incident-triage` and `breach-patterns` for AI-related incident response patterns; `csf-mapping` for the broader governance frame that AI RMF sits within.
8
9	## The NIST AI RMF — four functions
10
11	Just like the cybersecurity framework, the AI RMF organizes the work into functions. Same shape, different content.
12
13	\| Function \| What it covers \|
14	\|---\|---\|
15	\| Govern (GOV) \| Policy, accountability, roles, risk appetite, AI principles, board oversight, governance structures \|
16	\| Map (MAP) \| Context — what is the AI system, what does it do, who is impacted, what could go wrong, what are the legal / ethical constraints \|
17	\| Measure (MEAS) \| Evaluate the system — fairness, robustness, accuracy, explainability, privacy, security; quantitative + qualitative metrics \|
18	\| Manage (MAN) \| Treat the risks — mitigations, monitoring, incident response, decommissioning, ongoing review \|
19
20	The framework is voluntary but increasingly cited in contracts, RFPs, executive orders, and emerging regulations. Treat it as the lingua franca of AI risk.
21
22	## Workflow
23
24	### Step 1 — Inventory AI systems
25
26	Before assessment, build the inventory. Most organizations underestimate how much AI they actually deploy.
27
28	\| Category \| Examples \|
29	\|---\|---\|
30	\| First-party trained models \| Recommendation engines, fraud detection, churn prediction, internal ML pipelines \|
31	\| First-party LLM use \| Customer support chat, content generation, summarization, code generation, embeddings for search \|
32	\| Third-party AI features \| Stripe Radar (fraud), GitHub Copilot (code completion), Salesforce Einstein, Notion AI, Linear AI \|
33	\| Embedded AI in products you ship \| Suggested responses, smart defaults, AI sorting / ranking \|
34	\| AI in HR / hiring \| Resume screening, candidate matching, performance evaluation — high regulatory exposure \|
35	\| AI in customer-facing decisions \| Pricing, eligibility, content moderation, ad targeting — high regulatory exposure \|
36
37	For each, record: vendor (if any), training data source, deployment context, who it affects, the decision it informs, how decisions are reviewed.
38
39	### Step 2 — MAP: assess the context per system
40
41	For each AI system in the inventory, answer:
42
43	- Purpose — what is this system's stated goal? Does the actual deployment match?
44	- Stakeholders — who interacts with it, who is affected by its decisions, who is in a position to challenge those decisions?
45	- Legal / regulatory context — is this in scope for a specific law? (EU AI Act high-risk categories, US HUD fair-housing rules, EEOC for employment AI, FTC for unfair / deceptive practices, sector laws)
46	- Failure modes — what does "broken" look like? (Wrong answer, biased answer, hallucinated answer, slow answer, expensive answer, refused-to-answer-something-it-should, answered-something-it-should-not)
47	- Reversibility — when this system makes a wrong call, can the decision be undone? (Mortgage denial: hard to undo. Spam filter: easy)
48
49	### Step 3 — MEASURE: evaluate the system
50
51	The categories of evaluation, with the engineering hooks for each:
52
53	#### Accuracy / performance
54
55	- Test set evaluation — held-out data, not the training data
56	- Performance on slices of data, not just aggregate

briiirussellPublisher · 30 listings

Preview

briiirussell/cybersecurity-skills

# AI Risk Management — Beyond Security, the Whole Model Lifecycle

`prompt-injection` covers the AI security slice — attackers manipulating LLM inputs. This skill covers everything else risk-related about deploying AI / ML syst

The framing is NIST AI RMF 1.0 (released 2023) — the most widely-adopted voluntary framework — plus the regulatory layer (EU AI Act, US executive orders, sector

Cross-references: `prompt-injection` for prompt-injection / LLM-specific security attacks; `threat-modeling` for design-time AI risk modeling; `incident-triage`

Repobriiirussell/cybersecurity-skills

LicenseMIT

TypeSkills

CategorySecurity

UpdatedMay 2026

Repo stars320

First seenJul 22, 2026

People who liked this also viewed

6 picks

Type