cavendish
Do thisAssessedstrength · weak

Give citizen developers a paved road and a retention policy, not a review board

Review boards do not stop org slop; they slow the people who would have done it well. A paved road with a default retention window keeps the volume survivable.

Tier is not strength

Tier says how much we know. Strength says how hard we are telling you to act. Scored independently.

Evidence tierAssessed
Strengthweak
OwnerSKSam Kowalczyk
Last validated16 Apr 2026
Review by14 Aug 2026
Half-life120 days
Citations9
VerticalsCross-sector, Government, Retail & FMCG
decay20d overdue

Machine-readable target

{
  "configKey": "platform.citizen_dev.retention_days"
}

Nothing consumes it yet. Day two: findings ship as defaults into the gateway, routing config and skill library.

Body

Assessed from two validation runs and one graveyard entry, not from an experiment, which is why the tier is what it is. The observation that started the field: in one client, the number of agent-built internal tools went from around 40 to over 600 in five months, most of them unowned by month four (c-citizen-developers-org-slop-1).

g-citizen-dev-guardrails is the failed response. Template guardrails and a review board reduced the count of new tools by two-thirds and the count of useful new tools by about the same. The slop did not come from bad templates; it came from tools nobody deleted. The retention policy — anything without an owner and a use in ninety days is archived — did more than the board did, at no review cost (c-citizen-developers-org-slop-2).

This recommendation is weak because the retention-policy result is from one client and is confounded with a reorganisation that happened at the same time. It is stale because it has not been revalidated since April and the review date has passed; the field owner has it queued behind the eval work. Read it as a prior, not a finding, until the next run.

SKSigned Sam Kowalczyk · Research engineer · SDLC · 16 Apr 2026

What it rests on

In the enterprise tenants inventoried, most citizen-built agents are idle within ninety days and a sizeable minority duplicate an existing one.

Tried c-citizen-developers-org-slop-1
72%

Template guardrails reduce obvious failures but not sprawl, because sprawl is driven by discoverability and ownership, not quality.

Assessed c-citizen-developers-org-slop-2
64%

Field

Citizen Developers and Org Slop

Graveyard · refuted

Template guardrails stop org slop Guarded the template. Not the fork.