agentleFS
peterbamuhigirepeterbamuhigire/business-plan-skills/skills/meta-strategy/kaizen-improvement-system/SKILL.md

kaizen-improvement-system

Skill4 starsChanged 18 days agoMITView on GitHub

Use when auditing or improving the business-plan engine or any business plan, strategy, model, pitch, nonprofit plan, or execution framework it produces.

2 other projects use a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 8 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the local adoption plan and the portfolio standard.

    line 33
  2. Very specific:

    Create a P0/P1/P2 improvement backlog targeting 95/100.

    line 37
  3. Distinctive:

    Run a small Build-Measure-Learn or PDCA experiment.

    line 38
  4. Distinctive:

    Read and search are required.

    line 56
  5. Distinctive:

    If evidence, model, source, reviewer, or tool is unavailable, return the narrowest qualified result, mark the gap as not assessed, and do …

    line 60
  6. Distinctive:

    Never call a plan bankable or achievable without reconciled evidence, assumptions, finance, implementation, and risks.

    line 72
  7. Distinctive:

    Only after the capped baseline is recorded may the engine run the improvement cycle.

    line 78
  8. Distinctive:

    Run it twice when scope warrants: first for engine routing, models, references, validators, and handoffs.

    line 82
  9. Distinctive:

    If a market test raises enquiries but worsens fulfilment capacity, retain the result as mixed evidence, keep the capacity guardrail, revise …

    line 95
  10. Very specific:

    Every Kaizen cycle must begin with digital-research-engine source evaluation and source verification.

    line 99

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Kaizen Improvement System

Acknowledgement: Shared by Peter Bamuhigire, techguypeter.com.

<!-- dual-compat-start -->

Use When

  • Auditing this engine or a plan, model, strategy, pitch, or execution framework.
  • Converting assumptions, market feedback, KPI variance, or delivery lessons into a tested improvement.

Do Not Use When

  • The task is only a single-skill safety audit.
  • A current market, legal, tax, platform, or finance claim has not been routed to Digital Research and Chwezi where applicable.

Required Inputs

ArtefactSource/providerRequired?PurposeIf absent
Engine or product scope, audience, evidence pack, assumptions/model, current score, constraints, and target measuresProject and engineyesSet audit scope and improvement targetStop, mark unassessed, or request the missing evidence

Workflow

  1. Read the local adoption plan and the portfolio standard.
  2. Inventory routes, plan types, references, templates, models, examples, source registers, and release gates.
  3. Score doctrine, taxonomy, skill depth, applied proof, currency, output readiness, inclusivity, production fidelity, hygiene, routing, and integrity. Publish min(raw score, 65) and record blockers.
  4. For a product, test the thesis, customer or beneficiary logic, market evidence, operating model, ten-point local operational readiness, financial reconciliation, implementation capacity, risks, governance, and audience fit.
  5. Create a P0/P1/P2 improvement backlog targeting 95/100; each action needs owner, experiment, measure, evidence, and rollback.
  6. Run a small Build-Measure-Learn or PDCA experiment. If evidence fails, stop, recover the safe baseline, and revise the hypothesis.
  7. Re-run model, source, anti-slop, workbook, operational-readiness, and release gates as applicable; standardise successful learning and schedule the next review.

Outputs

ArtefactConsumerAcceptance condition
Capped audit, evidence gaps, blockers, 95/100 plan, experiment log, and standardisation recordPlanner, reviewer, and release ownerScores, owners, evidence, decisions, rollback, and re-audit date are explicit

Evidence Produced

EvidenceFormatAcceptance condition
Assumption ledger, model/source checks, experiment result, gate results, and before/after reviewMarkdown, workbook, or validator outputAnother reviewer can reproduce the decision and verify the change
<!-- dual-compat-end -->

Capability Contract

Read and search are required. Audits are read-only by default; edits require explicit authority and permission. Route current facts to Digital Research and finance doctrine to Chwezi.

Degraded Mode

If evidence, model, source, reviewer, or tool is unavailable, return the narrowest qualified result, mark the gap as not assessed, and do not certify bankability or readiness.

Decision Rules

ConditionActionFailure or risk avoided
Material assumption has no evidenceStop the claim and run a validation testFalse certainty
A test improves one metric but harms cash, mission, quality, or capacityReject, recover, or escalateLocal optimisation
A change passes gates and improves the target measureStandardise it and re-auditLost learning

Quality Standards

Never call a plan bankable or achievable without reconciled evidence, assumptions, finance, implementation, and risks. Never turn a book example into a client fact.

Mandatory 65-to-95 gate

The first pass is an initial analysis, not a certification: calculate the raw result, publish min(raw_score, 65), and keep every unassessed dimension and blocker visible. Only after the capped baseline is recorded may the engine run the improvement cycle. The cycle must target 95/100 with a named root cause, one reversible change, owner, measure, guardrail, stop/rollback rule, acceptance evidence, and re-audit date. A polished plan or book-derived idea is not a score increase until the evidence passes. Run it twice when scope warrants: first for engine routing, models, references, validators, and handoffs; then for the individual plan, model, pitch, or strategy.

Anti-Patterns

  • Adding initiatives without stopping or sequencing others. Fix: expose capacity trade-offs.
  • Treating a KPI as learning without a decision rule. Fix: define threshold, action, and counter-metric.
  • Writing a polished plan before testing assumptions. Fix: run a small validation experiment.
  • Hiding uncertainty in prose. Fix: label assumptions, gaps, confidence, and countercases.
  • Closing an improvement with no evidence. Fix: require a validated artefact or test.

Worked Example

If a market test raises enquiries but worsens fulfilment capacity, retain the result as mixed evidence, keep the capacity guardrail, revise the offer or process, and re-test before changing the plan.

Mandatory Digital Research currentness gate

Every Kaizen cycle must begin with digital-research-engine source evaluation and source verification. Record scope, dates, freshness class, support status, uncertainty, and review date for current market, legal, policy, technology, platform, finance, and lifecycle claims; quarantine unsupported claims as NOT_ASSESSED. Apply the portfolio Kaizen currentness gate.

References

  • Local adoption plan
  • Portfolio standard: resolve digital-research-engine through the global engine-routing table, then read docs/continuous-improvement/portfolio-kaizen-standard-2026-08.md.
  • skills/meta-strategy/meta-market-validation/
  • skills/meta-strategy/meta-living-plan-governance/
  • Book-driven commercial system and validation - whole-system trade-offs, validation, replication, GTM, cash, and currentness.
  • Marketing Plan Handbook operating loop - adaptive customer-first planning, segmentation, strategy-to-program alignment, forecasting, budgeting, metrics, and control.
  • Book-driven Kaizen Wave 3 - hypotheses, capacity/cash guardrails, experiments, dashboards, and responsible AI economics.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.