Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Risk Prioritization Strategy · yogsoth-ai | DecimalAI

yogsoth-ai/risk-prioritization

SkillScore68Maintained

Strategy: Action Priority matrix — classifies failure modes into H/M/L priority using severity-weighted scoring per AIAG-VDA 2019 Action Priority tables.

industryby yogsoth-ai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

QA & Test Engineering pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/yogsoth-ai-risk-prioritization/SKILL.md

decimalai skills pull yogsoth-ai-risk-prioritization --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/yogsoth-ai-risk-prioritization/SKILL.md https://app.decimal.ai/s/yogsoth-ai-risk-prioritization/SKILL.md

Raw file: /s/yogsoth-ai-risk-prioritization/SKILL.md ↗ · machine-readable summary at /s/yogsoth-ai-risk-prioritization.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

Runs

all versions · 30d

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 18d ago

—+24% tokens0% turns68% of whole cases pass with skill (15/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-16	✗→✓	▲ Improved	-61%	0%
case-01	✗→✓	▲ Improved	206%	0%
case-02	✗→✓	▲ Improved	-9%	0%
case-03	✗→✓	▲ Improved	12%	0%
case-06	✗→✓	▲ Improved	5%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/10/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

yogsoth-ai / de-anthropocentric-research-engine

skills/risk-prioritization/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 8/10/2026Version v1

Skill contents

SKILL.md

Risk Prioritization Strategy

Classifies failure modes into High/Medium/Low action priority using severity-weighted S/O/D scoring.

Method

Collect S/O/D scores from upstream FMEA analysis
action-priority-matrix applies AIAG-VDA Action Priority logic:

H (High): Mandatory action required — severity >= 8, or combined SO >= 36
M (Medium): Action recommended — moderate risk combinations
L (Low): Optional action — acceptable risk level

Rank all failure modes by priority class, then by RPN within class
Feed H-priority items to mitigation-design strategy

Budget Table

| Parameter | S | M | L | |---|---|---|---| | Failure modes classified | 8 | 20 | 40 | | Priority threshold tuning | fixed | 1 round | 2 rounds | | Sensitivity analysis | none | top-5 | full |

Orchestration

[S/O/D scores from upstream] → action-priority-matrix
  → priority-ranked failure catalog
  → [H-priority items] → mitigation-design

Subagents

action-priority-matrix (RPN computation and H/M/L classification)
severity-scoring (re-evaluation if needed)

Available Tactics

Optional, no fixed order; the final leaf is always a sop.

| Tactic | When to use | | --- | --- | | premortem-to-fmea-pipeline | Tactic: Pre-mortem rapid screening feeds high-risk items into full FMEA analysis. Bridges fast intuitive generation with systematic structured analysis. |

Available SOPs

Optional, no fixed order; the final leaf is always a sop.

| SOP | When to use | | --- | --- | | action-priority-matrix | Compute Risk Priority Number (RPN = S x O x D), classify failure modes into H/M/L action priority per AIAG-VDA tables. | | severity-scoring | Rate failure mode severity 1-10 based on end-effect impact. Follows AIAG-VDA severity scale calibrated for research artifacts. |

Illustrative Want to see what yogsoth-ai/risk-prioritization would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

9/3/2026, 22:27 UTC · 22 cases

Measured on 15 comparable cases — below the 22 needed to publish a lift, so this result is shown here but not on the registry card or leaderboard.

Lift

+55%

68% vs 14%

Agent turns

1.0 vs 1.0

Tokens

+24%

2,849.67 vs 2,305.73

fail → pass · 12pass → pass · 3fail → fail · 7

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-16 In an autonomous vehicle safety evaluation governed by the Large budget profile, how many rounds of priority threshold tuning should be carried out?	fail→pass	42,778	7,688	-82%	1	1	0%	2,157	836	-61%	0	0	—
case-01 I have a matrix of system failure modes with Severity, Occurrence, and Detection ratings from our infrastructure review. Could you process these scores, categorize each item into an action priority level (High/Medium/Low), and rank them within their respective priority classes? Finally, output the ranked catalog along with a list of the critical items that need immediate mitigation planning.	fail→pass	9,717	28,076	+189%	1	1	0%	1,563	4,779	+206%	0	0	—
case-02 In a medical device software FMEA review, failure mode FM-101 has a Severity rating of 8, Occurrence of 2, and Detection of 2. The project lead wants to categorize this as Low priority because its total Risk Priority Number is only 32. What action priority classification should be assigned to FM-101 under AIAG-VDA guidelines?	fail→pass	24,311	20,550	-15%	1	1	0%	3,689	3,346	-9%	0	0	—
case-03 For an API gateway component, failure mode FM-204 has Severity = 6, Occurrence = 6, and Detection = 1 (total score 36). A reviewer argues this should be Medium priority because Severity is below 8 and the total RPN is low. What action priority level applies to FM-204?	fail→pass	13,554	17,745	+31%	1	1	0%	2,383	2,669	+12%	0	0	—
case-04 During an e-commerce platform risk review, failure mode FM-302 is rated with Severity = 5, Occurrence = 4, and Detection = 3 (total RPN = 60). How should FM-302 be classified in the action priority matrix?	fail→fail	15,622	12,233	-22%	1	1	0%	1,903	1,865	-2%	0	0	—
case-05 A database backup procedure has failure mode FM-405 rated with Severity = 3, Occurrence = 2, and Detection = 2 (total RPN = 12). What is the appropriate action priority classification for this item?	pass→pass	15,076	11,039	-27%	1	1	0%	1,309	1,569	+20%	0	0	—
case-06 We have two High-priority failure modes from an industrial IoT audit: Failure Mode Alpha (Severity = 9, Occurrence = 2, Detection = 3, RPN = 54) and Failure Mode Beta (Severity = 8, Occurrence = 5, Detection = 3, RPN = 120). When compiling the ranked catalog within the High priority class, which failure mode should be ordered first?	fail→pass	16,472	12,339	-25%	1	1	0%	1,795	1,881	+5%	0	0	—
case-07 After completing AIAG-VDA action priority classification on 15 cloud deployment failure modes (resulting in 3 High, 7 Medium, and 5 Low priority items), which specific subset of failure modes should be passed to the mitigation design pipeline?	fail→fail	12,221	5,710	-53%	1	1	0%	1,804	1,454	-19%	0	0	—
case-08 An engineering team selects the Small execution profile from the risk budget framework for a quick audit. What is the maximum number of failure modes budgeted for classification under this profile?	fail→pass	27,648	2,277	-92%	1	1	0%	1,389	795	-43%	0	0	—
case-09 When running the risk classification workflow under the Medium resource allocation budget for a robotics pipeline, how many rounds of priority threshold tuning are conducted?	fail→pass	28,632	2,262	-92%	1	1	0%	1,274	830	-35%	0	0	—
case-10 An enterprise automotive safety team executes the risk prioritization process using the Large budget tier. What scope of sensitivity analysis is required for the evaluated catalog?	fail→pass	47,208	7,633	-84%	1	1	0%	1,049	907	-14%	0	0	—
case-11 For a Medium budget risk classification run on a payment gateway, what is the specified scope for performing sensitivity analysis on failure modes?	fail→pass	17,743	7,318	-59%	1	1	0%	1,910	828	-57%	0	0	—
case-12 A team is evaluating industrial sensor network risks using the Small budget profile. What sensitivity analysis protocol should be applied to the output catalog?	fail→pass	25,724	4,774	-81%	1	1	0%	3,130	1,018	-67%	0	0	—
case-13 What is the total failure mode processing capacity authorized when applying the Large budget allocation profile to an aerospace flight control evaluation?	fail→fail	53,882	16,356	-70%	1	1	0%	2,131	770	-64%	0	0	—
case-14 When conducting risk prioritization under the Medium budget profile for a telecom microservice suite, how many failure modes are budgeted for analysis?	fail→pass	36,584	2,412	-93%	1	1	0%	1,158	858	-26%	0	0	—
case-15 Under the Small budget allocation tier during a medical device software audit, how many threshold tuning iterations are specified for adjusting action priority cutoffs?	pass→pass	49,607	7,415	-85%	1	1	0%	1,336	842	-37%	0	0	—
case-17 In an avionics sensor bus review, failure mode FM-901 has a Severity rating of 10, Occurrence of 1, and Detection of 1 (RPN = 10). The team suggests that because occurrence is minimal and detection is immediate, it can be marked Low priority. What action priority level must be assigned?	fail→fail	25,225	21,704	-14%	1	1	0%	3,556	3,411	-4%	0	0	—
case-18 When evaluating failure mode FM-55 in a satellite power distribution system with Severity = 7, Occurrence = 4, and Detection = 3, what is the exact RPN value used for sorting items within its priority class?	fail→pass	12,642	10,087	-20%	1	1	0%	566	1,474	+160%	0	0	—
case-19 Before executing the action priority classification step for a satellite control system, what structured numerical ratings must be collected from the upstream analysis stage?	pass→pass	12,617	9,678	-23%	1	1	0%	1,220	1,341	+10%	0	0	—
case-20 Our risk evaluation flagged motor driver thermal overload (Severity = 8, Occurrence = 6, Detection = 2) as High priority on our motor drive board. Provide the detailed electrical engineering schematic and heat sink dimension calculations needed to redesign the motor drive board.	fail→fail	58,898	67,717	+15%	1	1	0%	2,819	4,406	+56%	0	0	—
case-21 We have provided 10,000 log records from our cloud edge routers over the past month. Compute the empirical Mean Time Between Failures (MTBF) and fit a Weibull failure rate distribution curve for the line card interface.	fail→fail	10,854	42,797	+294%	1	1	0%	1,939	8,743	+351%	0	0	—
case-22 Failure mode FM-88 is marked High priority due to a null pointer exception in auth_controller.py during token refresh. Write a suite of Python PyTest unit tests with mocks to patch and fix this code defect.	fail→fail	25,235	25,115	-0%	1	1	0%	4,230	4,593	+9%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 15 counted toward the lift figure. The other 7 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +55 percentage points is the difference between those two pass rates over the 15 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 13.6%
Pass rate with the skill: 68.2%
Cases attempted: 22
Cases counted toward lift: 15
Cases that regressed: 0
Skill version measured: v1
Run completed: 9/3/2026, 22:27 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-01

Prompt

I have a matrix of system failure modes with Severity, Occurrence, and Detection ratings from our infrastructure review. Could you process these scores, categorize each item into an action priority level (High/Medium/Low), and rank them within their respective priority classes? Finally, output the ranked catalog along with a list of the critical items that need immediate mitigation planning.

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run68

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns-3% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 17 days ago

Other measured skills in the registry, with their headline benchmark lift.