regression

DE//AGENT currently lists 4 resources under regression: 1 tutorial, 1 skill, 1 hook. Everything below is derived from the live registry — counts, capabilities, verification state and security posture come from the curated records, never from marketing copy.

Also present: 1 runbook.

Documented learning objectives across these resources:

• Capture cost-relevant metrics from a Spark run — via Spark Cost Regression Guard
• Store and compare a baseline — via Spark Cost Regression Guard
• Explain which stage caused a regression — via Spark Cost Regression Guard
• Diff shuffle write and input bytes run-over-run — via Attribute a 4x Spark cost regression
• Attribute a cost regression to one stage — via Attribute a 4x Spark cost regression
• Gate deploys on shuffle/scan metrics — via Attribute a 4x Spark cost regression

These resources map to stage: Reliability (4). Follow the four-phase roadmap to place this topic inside a progression instead of learning it in isolation.

Every entry, its shape and its trust state — sorted by the registry's composite score. This table is the honest version of a marketing comparison: no sponsored placements, no unverifiable claims.

How scores are computed: source quality score.

Highest composite score first:

1. Spark Cost Regression Guard score 37 — advanced
2. Spark job suddenly costs 4x score 36.75 — advanced
3. Attribute a 4x Spark cost regression score 34.15 — intermediate

Scores rank provenance, completeness and trust — read how verification works before adopting anything.

Reach for these with production experience: Spark Cost Regression Guard, Spark job suddenly costs 4x, Spark Cost Check.

Security-flagged entries: Spark Cost Regression Guard (2 flags), Spark job suddenly costs 4x (1 flag), Spark Cost Check (2 flags). Flags mean the record declares capabilities like credential use or write access — read the security posture on each page before connecting.

• Spark Cost Regression Guard — Compares a job's shuffle/spill/scan metrics against a stored baseline and fails the change if cost-per-run regresses beyond a threshold, with the diff explained. details
• Spark job suddenly costs 4x — One job's compute cost quadrupled without an obvious functional change — attribute it and reverse the regression. details
• Attribute a 4x Spark cost regression — One job's compute cost quadrupled overnight with no functional change. Run the 4x runbook: diff shuffle and input bytes run-over-run, find the guilty stage, reverse the regression … details
• Spark Cost Check — Fail a build when a Spark job's cost metrics regress past a threshold. details

Real install commands from the registry — read the source before running any of them:

npx de-agent-platform add spark-cost-regression-guard

Spark Cost Regression Guard — review first

Adjacent topics: cost (4 shared), ci (2 shared).

Study path: the roadmap places this topic in a progression · the scoring model explains the numbers · the trust rule catalog shows every security check these entries passed through.