Commit graph

5 commits

Author SHA1 Message Date
Alex Gorevski
44725da08c perf(ci): reduce GitHub Actions costs ~60-65% across all workflows
Analysis of Feb 17 data showed 400+ workflow runs/day consuming ~398 billable minutes (~200 hours/month projected). Implemented targeted optimizations:

High-impact changes:

- sec-audit.yml: add path filters (Cargo.toml, src/**, crates/**, deny.toml); skip docs-only PRs

- test-benchmarks.yml: move from every-push-to-main to weekly schedule; retention 30d -> 7d

- pub-docker-img.yml: tighten PR smoke build path filters to Docker-specific files only

- sec-codeql.yml: reduce from twice-daily (14 runs/week) to weekly

Medium-impact changes:

- ci-run.yml: merge lint + lint-strict-delta into single job; drop --release from smoke build

- feature-matrix.yml: remove push trigger (weekly-only); remove redundant cargo test step

- dependabot.yml: monthly instead of weekly; reduce PR limits from 11 to 5/month; group all deps

Runner cost savings:

- Switch 6 lightweight API-only workflows to ubuntu-latest (PR Labeler, Intake, Auto Responder, Check Stale, Check Status, Sync Contributors)

- pr-check-status.yml: reduce from every 12h to daily

New files:

- docs/ci-cost-optimization.md: comprehensive analysis and revised architecture documentation

- scripts/ci/fetch_actions_data.py: reusable GitHub Actions cost analysis script

Estimated impact: daily billable minutes ~400 -> ~120-150 (60-65%% reduction), monthly hours ~200 -> ~60-75, Dependabot PRs ~44/month -> ~5 (89%% reduction)
2026-02-18 21:14:47 -08:00
Will Sarg
5afff7b6d7
ci(cache): migrate rust cache actions to useblacksmith (#706) 2026-02-17 18:05:31 -05:00
Will Sarg
e9f619a83f
ci: cut default PR runtime to build/security core checks (#669) 2026-02-17 16:20:13 -05:00
Will Sarg
42f1d40f1f
fix(ci): unblock dependabot dependency PR checks (#658) 2026-02-17 15:51:07 -05:00
Alex Gorevski
7ea23da0c6
ci: add feature matrix testing workflow (#634)
Problem: CI only tests the default feature set. The codebase defines
multiple Cargo features (hardware, browser-native, sandbox-landlock,
sandbox-bubblewrap, probe, rag-pdf) behind conditional compilation.
Feature-gated code can silently break without CI coverage.

Solution: Add a dedicated feature-matrix workflow that tests key
feature combinations in a matrix strategy:
- --no-default-features (bare minimum compiles)
- --all-features (everything together)
- --no-default-features --features hardware (isolated hardware)
- --no-default-features --features browser-native (isolated browser)

Each combination runs both cargo check and cargo test. The
workflow triggers on Cargo.toml/lock/src changes and weekly schedule.

Testing: Validated YAML syntax and matrix expansion logic. Actual
feature compilation will be verified by CI on first run.

Ref: zeroclaw-labs/zeroclaw#618 (item 2 — Feature Matrix Testing)

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-02-17 15:10:02 -05:00