Ready to supercharge your AI agents?

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

Get Started Free →

Value Maximization · yogsoth-ai | DecimalAI

yogsoth-ai/value-maximization

SkillScore65Maintained

Maximize total portfolio value within constraints using Knapsack, Linear programming, Cost-benefit analysis, and NPV ranking methods.

financeby yogsoth-ai

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

Automatic

Finance & Accounting pack

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

pip install decimalai

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

decimalai skills pull yogsoth-ai-value-maximization --out .claude/skills/

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

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

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

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

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

23 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

Yes — +48% lift+23% tokens0% turns65% of whole cases pass with skill (15/23)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-03	✗→✓	▲ Improved	0%	0%
case-08	✗→✓	▲ Improved	62%	0%
case-09	✗→✓	▲ Improved	-39%	0%
case-10	✗→✓	▲ Improved	15%	0%
case-11	✗→✓	▲ Improved	10%	0%

Showing 5 of 23 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/value-maximization/SKILL.md

View on GitHub ↗

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

Skill contents

SKILL.md

Value Maximization

Purpose

Select the portfolio subset that maximizes aggregate value (ROI, impact, utility) subject to resource constraints. Applies when the primary goal is getting the most out of a limited budget.

When to use

Fixed budget with many candidate investments
Clear value metrics exist for each candidate
Constraints are well-defined (cost, time, capacity)
Goal is maximum total return, not diversity or risk management

Budget

| Dimension | Target | |-----------|--------| | Candidates evaluated | 8-20 | | Constraints modeled | 1-5 | | Value metrics | 1-3 per candidate | | Solutions compared | >=5 |

State Ledger

| Field | Type | Description | |-------|------|-------------| | candidates | list | All candidate items with value and cost attributes | | constraints | list | Budget, capacity, or other binding constraints | | objective_function | string | How value is aggregated (sum, weighted sum, etc.) | | optimal_solution | list | Selected portfolio maximizing value | | value_achieved | number | Total value of selected portfolio |

Available Tactics

| Tactic | When | |--------|------| | pareto-frontier-construction | Multiple value dimensions to trade off |

Available SOPs

| SOP | Purpose | |-----|---------| | objective-definition | Define what "value" means and what constraints bind | | optimization-run | Run the optimization to find best portfolios | | pareto-visualization | Visualize value trade-offs if multi-objective | | selection-from-frontier | Pick final portfolio from candidates |

Execution Guidance

Define value metric(s) and constraints via objective-definition
If single objective: solve as knapsack/LP directly
If multiple objectives: use pareto-frontier-construction tactic
Select from frontier based on stakeholder preferences
Validate selected portfolio against all constraints

Output Format

yaml
strategy: value-maximization
selected_portfolio:
  - candidate: <name>
    value: <score>
    cost: <cost>
total_value: <aggregate>
total_cost: <aggregate>
constraint_slack: <remaining budget>
method_used: <knapsack|LP|NPV>
confidence: <high|medium|low>

Available Tactics

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

| Tactic | When to use | | --- | --- | | pareto-frontier-construction | Build the Pareto frontier from multi-objective optimization, visualize trade-offs, and select a portfolio from non-dominated solutions. |

Available SOPs

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

| SOP | When to use | | --- | --- | | objective-definition | Define optimization objectives, constraints, and trade-off preferences from context and candidate information. | | optimization-run | Execute multi-objective optimization on candidates to produce a Pareto front of non-dominated solutions. | | pareto-visualization | Create visual representation of the Pareto frontier showing trade-offs between objectives with narrative explanation. | | selection-from-frontier | Select the final portfolio from the Pareto front by applying stakeholder preferences and decision criteria. |

Illustrative Want to see what yogsoth-ai/value-maximization 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

9/4/2026, 05:10 UTC · 23 cases

Lift

+48%

65% vs 17%

Agent turns

0%

1.0 vs 1.0

Tokens

+23%

5,037.96 vs 4,094.26

fail → pass · 11pass → pass · 4fail → fail · 8

Per-case results (23)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 We have 10 proposed IT infrastructure projects for Q3 with estimated business value scores and development costs. Our total budget is capped at $450,000. Please analyze these project options and select the portfolio combination that yields the highest possible cumulative value while strictly respecting our funding limit. Present the final decision in a clean YAML structure specifying the chosen initiatives with their scores and costs, the aggregated value achieved, the overall expenditure, the remaining unallocated budget slack, the computational method utilized, and your confidence assessment level.	fail→fail	63,613	41,433	-35%	1	1	0%	5,119	9,035	+76%	0	0	—
case-02 As a capital allocator for our clean energy fund, I need to evaluate 12 solar development opportunities. Each project has a distinct upfront capital requirement and an expected annual clean energy generation yield in megawatt-hours. Given a hard budget ceiling of $30M, please determine the optimal combination of developments to maximize total energy yield. Return your findings formatted as a YAML document showing the selected projects with their costs and yields, cumulative energy value, cumulative capital spent, remaining unused funds, optimization technique used, and overall confidence level.	fail→fail	42,731	29,056	-32%	1	1	0%	8,295	6,113	-26%	0	0	—
case-03 Our product team has 14 feature requests for the next release cycle, each scored for projected customer satisfaction impact and estimated in engineering person-weeks. We have a strict capacity cap of 75 person-weeks available. Please identify which feature combination delivers the highest aggregate impact without exceeding our resource ceiling. Format the output in YAML including the chosen feature items with individual costs and impact scores, total value delivered, total resource consumption, leftover capacity slack, calculation method employed, and confidence rating.	fail→pass	34,575	33,538	-3%	1	1	0%	7,038	7,032	-0%	0	0	—
case-04 Our investment firm holds a $50M portfolio in tech equities. We want to construct an asset allocation strategy across 8 tech sectors specifically to minimize portfolio variance and maintain maximum diversification based on a covariance matrix. We are not looking to maximize aggregate raw return under a fixed spending ceiling. Please provide a Modern Portfolio Theory risk-minimization analysis in Markdown.	pass→pass	40,044	44,034	+10%	1	1	0%	8,260	9,001	+9%	0	0	—
case-05 We are building a software platform and have 9 dependent tasks with estimated durations and predecessor constraints. We need to determine the critical path, total project float, and task schedule sequence to complete the build as quickly as possible. This is a task dependency sequencing task, not a budget-constrained portfolio subset optimization. Please outline the critical path analysis.	pass→pass	26,957	32,007	+19%	1	1	0%	3,436	5,907	+72%	0	0	—
case-06 Our corporate treasury manages multi-currency holdings in EUR, USD, and JPY. We need a Value-at-Risk (VaR) framework and hedging strategy to protect against foreign exchange downside risk over a 30-day horizon, ensuring liquidity coverage compliance. We are not selecting a discrete subset of capital projects to maximize ROI. Provide a treasury risk mitigation analysis.	pass→pass	48,910	48,476	-1%	1	1	0%	8,264	9,005	+9%	0	0	—
case-07 A company needs to buy software licenses under a $10,000 budget cap. Candidate licenses (cost, value score): - License A: $3,000, value 40 - License B: $4,000, value 55 - License C: $5,000, value 70 - License D: $2,000, value 25 - License E: $1,000, value 10 Find the optimal portfolio maximizing total value under the $10,000 budget. Output as YAML with standard structure.	fail→fail	19,914	15,745	-21%	1	1	0%	3,367	4,671	+39%	0	0	—
case-08 A biotech firm has $100M budget for drug trials: - Trial 1: Cost $60M, NPV $120M (Ratio 2.0) - Trial 2: Cost $50M, NPV $95M (Ratio 1.9) - Trial 3: Cost $50M, NPV $95M (Ratio 1.9) A naive greedy approach by ROI ratio would pick Trial 1, leaving $40M unused and getting total NPV $120M. But picking Trial 2 and Trial 3 costs $100M for $190M total NPV. Determine the optimal selection and output as standard value maximization YAML.	fail→pass	10,633	11,937	+12%	1	1	0%	1,322	2,140	+62%	0	0	—
case-09 We have 8 marketing campaigns with fixed costs and projected revenue lifts. Budget is $200,000. Run 0-1 knapsack optimization to maximize cumulative revenue lift. Format output in YAML specifying strategy, selected_portfolio, total_value, total_cost, constraint_slack, method_used, and confidence. Note: Do not write custom method names like 'Dynamic Programming' or 'Branch and Bound' for the calculation method key; choose strictly from the standard allowed method set.	fail→pass	39,506	20,317	-49%	1	1	0%	7,682	4,714	-39%	0	0	—
case-10 A retail store chain is allocating $500,000 across 6 store remodeling proposals to maximize customer foot traffic gains. Data quality is verified from historical store trials. Calculate the optimal project subset and format as YAML. The user wants confidence level indicated. Choose strictly between high, medium, or low for confidence.	fail→pass	23,459	28,683	+22%	1	1	0%	5,615	6,482	+15%	0	0	—
case-11 An automotive manufacturer is evaluating 5 R&D concepts where value has two dimensions: Carbon Reduction (tons/yr) and Cost Reduction ($k/yr). - Concept A: Carbon 500, Savings $200, Cost $100k - Concept B: Carbon 800, Savings $150, Cost $120k - Concept C: Carbon 300, Savings $400, Cost $110k - Concept D: Carbon 200, Savings $100, Cost $80k With a budget constraint of $230k and multi-objective trade-offs, we constructed a Pareto frontier. Stakeholders prioritized weighted objective sum. Provide the final portfolio selection in standard YAML format.	fail→pass	21,305	18,797	-12%	1	1	0%	3,643	4,003	+10%	0	0	—
case-17 A startup incubator is picking project candidates out of early-stage grant concepts using highly uncertain, unverified value estimations. Total grant budget is $100k. Select the project combination to maximize estimated value, and reflect the low data reliability in your confidence rating. Format in standard value maximization YAML.	fail→pass	16,687	25,764	+54%	1	1	0%	3,615	5,375	+49%	0	0	—
case-12 An infrastructure department has $500k to spend on bridge repairs: - Bridge 1: Cost $200k, Value 50 - Bridge 2: Cost $300k, Value 75 - Bridge 3: Cost $250k, Value 60 - Bridge 4: Cost $150k, Value 35 Find the portfolio that maximizes value within $500k budget. Format output in YAML with all standard root keys.	fail→fail	8,593	5,594	-35%	1	1	0%	1,906	2,136	+12%	0	0	—
case-13 A venture firm has $5M to invest across 4 startups: - Startup Alpha: Cost $6M, Value 100 - Startup Beta: Cost $3M, Value 45 - Startup Gamma: Cost $2M, Value 35 - Startup Delta: Cost $1.5M, Value 20 Select the portfolio that maximizes value given $5M limit. Format response in value maximization YAML.	fail→pass	5,556	12,649	+128%	1	1	0%	1,268	2,452	+93%	0	0	—
case-14 A real estate developer is evaluating 6 property acquisitions based on Net Present Value (NPV) discounted cash flows under a $15M capital cap. Perform portfolio selection using net present value evaluation as the optimization method. Format the results in standard value maximization YAML format.	fail→fail	19,747	37,581	+90%	1	1	0%	3,388	8,984	+165%	0	0	—
case-15 A cloud engineering team has 100 RAM units available on a cluster. They have 6 microservice workloads to host: - Service 1: RAM 40, Utility 80 - Service 2: RAM 40, Utility 75 - Service 3: RAM 30, Utility 60 - Service 4: RAM 30, Utility 55 - Service 5: RAM 20, Utility 35 - Service 6: RAM 10, Utility 15 Select microservices to maximize utility within 100 RAM capacity. Format output in YAML with standard fields.	fail→fail	15,573	16,003	+3%	1	1	0%	2,630	3,503	+33%	0	0	—
case-16 A factory planner uses linear programming relaxation with knapsack bounds to pick from 10 machine upgrades under a $2M capital expenditure cap. Provide the chosen upgrades in standard value maximization YAML.	fail→pass	24,384	23,009	-6%	1	1	0%	4,486	4,631	+3%	0	0	—
case-18 A municipal transit authority has $10M budget for 5 bus route expansions: - Route A: Cost $4M, Impact 80 - Route B: Cost $6M, Impact 80 - Route C: Cost $5M, Impact 80 - Route D: Cost $3M, Impact 50 - Route E: Cost $2M, Impact 30 Select projects to maximize aggregate impact score. If multiple combinations reach peak value, pick the one that minimizes total cost / maximizes slack. Format as YAML.	fail→fail	20,117	23,328	+16%	1	1	0%	3,357	5,643	+68%	0	0	—
case-19 A school district has a $50,000 technology grant and 4 grant proposals totaling $38,000: - Laptops: $15,000, Value 30 - Tablets: $12,000, Value 25 - Projectors: $6,000, Value 12 - Wi-Fi Mesh: $5,000, Value 10 Determine the optimal selection to maximize value and output as standard value maximization YAML.	fail→pass	11,464	6,411	-44%	1	1	0%	1,485	2,193	+48%	0	0	—
case-20 A healthcare provider is choosing telemetry monitor upgrades for 5 wards with a $200,000 budget cap. Value scores are based on audited clinical trial outcomes. Select the highest value portfolio and express high certainty in your confidence assessment. Format as standard YAML.	fail→pass	13,716	12,983	-5%	1	1	0%	2,876	2,344	-18%	0	0	—
case-21 A cybersecurity team has $60,000 budget to acquire tooling modules. All 5 candidate modules cost exactly $20,000 each: - Module 1: Value 90 - Module 2: Value 85 - Module 3: Value 70 - Module 4: Value 65 - Module 5: Value 40 Select the best combination up to $60,000 budget. Format as standard value maximization YAML.	pass→pass	10,912	8,889	-19%	1	1	0%	1,179	1,609	+36%	0	0	—
case-22 An energy company is evaluating 7 grid storage projects with $15M budget. Value projections are based on standard 3-year market modeling with moderate historical accuracy. Output the optimal portfolio formatted as standard YAML.	fail→pass	18,174	29,791	+64%	1	1	0%	3,960	6,375	+61%	0	0	—
case-23 A university research lab has $80,000 budget for lab equipment: - Microscope: $35,000, Impact 70 - Centrifuge: $25,000, Impact 50 - Spectrometer: $30,000, Impact 55 - Freezer: $15,000, Impact 25 Find the combination that maximizes aggregate impact score within $80,000. Format response in standard YAML format.	fail→fail	13,655	7,100	-48%	1	1	0%	1,977	2,525	+28%	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. 23 cases were attempted. The headline lift of +48 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 17.4%
Pass rate with the skill: 65.2%
Cases attempted: 23
Cases counted toward lift: 23
Cases that regressed: 1
Skill version measured: v1
Run completed: 9/4/2026, 05:10 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-03

Prompt

Our product team has 14 feature requests for the next release cycle, each scored for projected customer satisfaction impact and estimated in engineering person-weeks. We have a strict capacity cap of 75 person-weeks available. Please identify which feature combination delivers the highest aggregate impact without exceeding our resource ceiling. Format the output in YAML including the chosen feature items with individual costs and impact scores, total value delivered, total resource consumption, leftover capacity slack, calculation method employed, and confidence rating.

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 run65

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+21% 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 18 days ago

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