agentsclimarketplace
Sign in

Process optimizer

Skill SamyakJhaveri/loam/cultivation/marketplace/business-process/skills/process-optimizer

Analyzes existing workflows and processes to find inefficiencies, bottlenecks, and optimization opportunities. Goes beyond automation to improve the process itself. Use when user says "optimize my process", "find bottlenecks", "why is this slow", "improve this workflow", "process audit", or "efficiency review".From its SKILL.md

Install
npx -y skills add SamyakJhaveri/loam --skill process-optimizer

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

4.2 KB, 954 tokens by cl100k_base, as published. Nobody here has run it

Process Optimizer — Find and Fix Bottlenecks

Analyzes workflows not just for automation opportunities, but for process improvement — eliminating unnecessary steps, reducing handoffs, parallelizing work, and fixing the root cause of inefficiency.

How It Works

Step 1: Receive the Process

Accept a workflow in any format:

  • Output from /workflow-mapper
  • Numbered step list
  • Free-form description
  • "Here's how we handle [X]..."

Step 2: Run the 5-Layer Analysis

For each step in the process, evaluate against these 5 lenses:

1. Necessity Check: Does this step need to exist at all?

  • Is it redundant with another step?
  • Is it legacy (existed for a reason that no longer applies)?
  • Would removing it break anything?

2. Sequence Check: Is this step in the right place?

  • Could it happen earlier (preventing downstream waste)?
  • Could it happen in parallel with other steps?
  • Is it blocking something it shouldn't be?

3. Handoff Check: Does this step involve unnecessary handoffs?

  • Does work pass between people/tools unnecessarily?
  • Could one person/tool handle adjacent steps?
  • Is there a wait time between handoff and pickup?

4. Error Check: Where does this step fail?

  • What are the common errors or rework triggers?
  • Is there a validation step, or do errors propagate downstream?
  • What's the cost of failure at this step?

5. Automation Check: Can technology handle this?

  • Same as the GREEN/YELLOW/RED framework from workflow-mapper
  • But also considers partial automation (AI assists, human decides)

Step 3: Generate the Optimization Report

# Process Optimization: [Process Name]

## Current State
- **Steps**: [X]
- **Estimated time per occurrence**: [X min/hours]
- **Handoffs**: [X between people/tools]
- **Known pain points**: [from user's description]

## Findings

### Eliminate (Steps That Don't Need to Exist)
| Step | Reason to Remove | Time Saved |
|------|-----------------|------------|
| [Step X] | [Redundant with Step Y] | [X min] |

### Resequence (Steps in the Wrong Order)
| Step | Current Position | Recommended Position | Why |
|------|-----------------|---------------------|-----|
| [Step X] | After Step Y | Before Step Y | [Prevents Z rework] |

### Parallelize (Steps That Can Run Simultaneously)
| Steps | Currently | Recommended | Time Saved |
|-------|-----------|-------------|------------|
| [X + Y] | Sequential (20 min) | Parallel (10 min) | 10 min |

### Reduce Handoffs
| Handoff | Current Flow | Recommended | Why |
|---------|-------------|-------------|-----|
| [Step X → Y] | [Person A → Person B] | [Person A handles both] | [Eliminates 2hr wait] |

### Automate
[Use GREEN/YELLOW/RED framework — reference /workflow-mapper output if available]

## Optimized Process
[Rewritten process with all improvements applied — numbered steps, clean and clear]

## Impact Summary
| Metric | Before | After | Improvement |
|--------|--------|-------|-------------|
| Total steps | [X] | [Y] | -[Z] steps |
| Time per occurrence | [X min] | [Y min] | -[Z]% |
| Handoffs | [X] | [Y] | -[Z] |
| Error-prone steps | [X] | [Y] | -[Z] |
| **Time saved per week** | — | **[X hours]** | — |

Step 4: Implementation Recommendations

Provide a prioritized list:

  1. Do now (no-cost, no-tool changes — just resequence/eliminate)
  2. Do this week (simple automation or tool configuration)
  3. Do this month (requires building, setup, or team training)

Rules

  • Don't just find things to automate. The biggest wins often come from REMOVING steps, not automating them.
  • Question every handoff. Each handoff introduces wait time and context loss.
  • Before and after must be concrete. Show the math.
  • Respect institutional knowledge. If a step looks unnecessary but exists for regulatory, legal, or safety reasons, flag it as "verify with stakeholder" rather than recommending removal.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most performance cost skills give in 954 tokens

Counted across 803 of the 1,058 authors here whose files we hold, read 2026-08-07

  • Keep skill files under 500 lines or tokensin 82 of 803, across 16 files
  • Use imperative form in instructionsin 80 of 803, across 9 files
  • Draft assertions while test runs are in progressin 75 of 803, across 9 files
  • Create two to three realistic test promptsin 74 of 803, across 9 files
  • Write skill descriptions to be pushyin 72 of 803, across 7 files
  • Save test cases to evals JSONin 72 of 803, across 6 files
  • Ask questions about edge cases and input formatsin 72 of 803, across 7 files
  • Save timing data immediately when runs completein 70 of 803, across 5 files
  • Include all trigger conditions in the skill descriptionin 69 of 803, across 3 files
  • Launch all test runs in a single turn or simultaneouslyin 69 of 803, across 3 files
  • Capture intent before writing a skillin 67 of 803, across 1 file
  • Import directly instead of barrel filesin 52 of 803, across 15 files

Said here and by no other author read

  • evaluate each step against five specific lenses
  • check if each step needs to exist
  • question every handoff for necessity
  • identify common errors and rework triggers
  • provide prioritized implementation recommendations
  • show concrete before and after math

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,614. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.