testland/chaos-results-reporter
Aggregates chaos drill verdicts over time into a resilience trend report - per-experiment hypothesis-held / blast-radius / time-to-detect / time-to-recover, degradation trends across runs, action items, and a stakeholder summary. Use when a team has completed one or more chaos drills and needs a structured trend report showing whether resilience is improving, degrading, or stable across iterations.
74
93%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Low
Low-risk findings worth noting
drill-fields-and-signals.mdreferences/
Drill report field extraction and degradation signals
Reference tables for chaos-results-reporter. Step 1 uses the field-extraction
map to parse each drill report into a normalized row; Step 4 uses the
degradation-signals catalog to flag findings across runs.
Field extraction map
For each completed drill report, extract these fields from the listed source
locations. If any required field is missing from a report, flag that report as
INCOMPLETE in the aggregate table and skip it from trend calculations - do not
guess or interpolate missing values.
| Field | Source location in drill report |
|---|---|
experiment_id | Header: Chaos drill report - <id> |
date | Start: timestamp, truncated to date |
experiment_type | Experiment: field |
hypothesis | Steady-state hypothesis from experiment YAML |
verdict | Verdict: field: PASSED, ABORTED, FAILED |
blast_radius_observed | Peak error rate, Peak affected replicas, Peak latency p99 |
blast_radius_bound | Blast-radius bound: field |
time_to_detect | Time from injection start to first abort criterion breach (or "n/a" if not aborted) |
time_to_recover | Recovery time: field |
abort_reason | Abort reason: field (empty if verdict is PASSED) |
Degradation signals catalog
Scan all experiments for these signals. Emit each match as a named finding; each HIGH or MEDIUM finding becomes an action item in Step 5.
| Signal | Condition | Severity |
|---|---|---|
| Repeated blast-radius breach | Same experiment type aborted for the same abort reason in 2+ consecutive runs | HIGH |
| No TTR improvement after fix | ABORTED run was followed by a code change (per git log or team note), but TTR did not improve in the next run | HIGH |
| Widening blast radius | Blast-radius trend is WIDENING for any experiment type | MEDIUM |
| Declining held rate | Held rate dropped more than 20 percentage points between first and second half of run history | MEDIUM |
| Single data point | Any experiment type has only one run | LOW (informational) |