Review performance and efficiency
Assess one repository, a repository group, or an existing fleet evidence bundle. Prefer measurements and label static hot-path concerns as potential constraints.
Run the lens
- Read the standalone lens protocol completely and prepare the supplied scope with lens ID
performance-efficiency.
- Read the performance and efficiency checks completely. Apply every check family relevant to workload shape, application type, build outputs, and available measurements.
- Trace work, data movement, allocation, and blocking across likely hot paths. Classify measured or well-supported outcomes as strengths or findings; place missing workload and measurement evidence in unknowns and coverage.
- Follow the protocol to account for every contract-registered check family, validate the result, ingest it, and hand back the canonical artifact path.
For raw collection, builds, benchmarks, or profilers, read and follow the safety model. Run repository-controlled measurement only through authorized deep-mode adapters.
Read scanner adapters before interpreting profiler, bundle, or build-tool output.
Boundary
Own latency, throughput, capacity, resource use, frontend loading, and build cost. Route correctness under exhaustion or failure to software reliability and production signal coverage to operability.
Completion
Complete when: every in-scope repository has an exact commit or explicit unavailable state, every applicable check family is accounted for, every claim has bounded sanitized evidence at an appropriate confidence, gaps remain visible in unknowns and coverage, and the shared result validator succeeds.
1---2name: review-performance-efficiency3description: Performance and efficiency review for hot paths, I/O, database access, batching, caching, frontend loading, resource use, build cost, and scaling constraints. Use for latency, throughput, bundle-size, or efficiency questions.4---56# Review performance and efficiency78Assess one repository, a repository group, or an existing fleet evidence bundle. Prefer measurements and label static hot-path concerns as potential constraints.910## Run the lens11121. Read the [standalone lens protocol](../audit-software-fleet/references/standalone-lens.md) completely and prepare the supplied scope with lens ID `performance-efficiency`.132. Read [the performance and efficiency checks](references/checks.md) completely. Apply every check family relevant to workload shape, application type, build outputs, and available measurements.143. Trace work, data movement, allocation, and blocking across likely hot paths. Classify measured or well-supported outcomes as strengths or findings; place missing workload and measurement evidence in unknowns and coverage.154. Follow the protocol to account for every contract-registered check family, validate the result, ingest it, and hand back the canonical artifact path.1617For raw collection, builds, benchmarks, or profilers, read and follow the [safety model](../audit-software-fleet/references/safety-model.md). Run repository-controlled measurement only through authorized deep-mode adapters.1819Read [scanner adapters](../audit-software-fleet/references/scanner-adapters.md) before interpreting profiler, bundle, or build-tool output.2021## Boundary2223Own latency, throughput, capacity, resource use, frontend loading, and build cost. Route correctness under exhaustion or failure to software reliability and production signal coverage to operability.2425## Completion2627**Complete when:** every in-scope repository has an exact commit or explicit unavailable state, every applicable check family is accounted for, every claim has bounded sanitized evidence at an appropriate confidence, gaps remain visible in unknowns and coverage, and the shared result validator succeeds.
kenlck