[JacobStephens2]

Summary

Cron jobs, Launch directory discovery and Console starts now obtain their independent facts through one owned interpretation. Existing recognition rules and visible results are preserved.

Cron command text
└── cronjobs::Interpretation
    ├── entry recognition + Command kind → Cron jobs inventory
    ├── literal Launch facts + output log → filesystem/GitHub-origin resolution
    └── parsed CLI candidate → Console supported-Pass/request validation

Adds the Cron job glossary definition, updates the domain-to-code index, and characterizes the same fake crontab through CLI and owning-machine Console projections.

Closes #2008

Evidence

Execution evidence

• Before: cargo test --bin thirdshift cronjobs::command::tests::interpretation_preserves_independent_cron_job_facts failed with error[E0433]: cannot find type Interpretation in this scope at the new interface seam.
  After: the same command passed: 1 passed; 0 failed. The final interpreter suite (cargo test --bin thirdshift cronjobs::command::tests) passed all 3 tests.
• The shared-crontab characterization passed against the original callers: cargo test --test console cron_job_projections_agree_without_conflating_kind_directory_and_start → 1 passed; 0 failed. After caller migration, cargo test --test console cronjobs:: passed all 13 tests, including that characterization, Fleet/legacy metadata, scope checks and malformed/display-only entries.
• cargo check --bin thirdshift, cargo fmt --check, and CARGO_BUILD_JOBS=2 cargo clippy --all-targets -- -D warnings passed.
• Final validation passed: CARGO_BUILD_JOBS=2 NEXTEST_TEST_THREADS=4 RUST_TEST_THREADS=3 /usr/bin/time -p scripts/affected-tests main. Site: 68 tests passed (127.82s); Console browser: 197 tests passed (273.31s); full Rust: 4,214 tests passed, 97 skipped. The selector's cargo nextest run took 2,936.34s, including compilation (2m 36s); total selector wall time was 3,442.21s. The Rust run used four test slots on this host.
• All required checks passed in CI, alongside the x86_64 and ARM Linux release builds.

Test seams

• Cron job interpretation: recognition, Command kind, literal Launch facts, selected output log and parsed CLI candidate through one owned interpretation.
• Existing CLI and owning-machine Console interfaces: the same fake crontab projected into inventory, Launch directories and supported Run now requests.

Review coverage

Independent Standards and Spec reviewers each read all seven changed files against refs/remotes/origin/main, pinned to 05171b3c8b80296aed7808c6f93bf5ac7bdf1187. Standards: 0 findings. Spec: 0 findings. Changed files left unread: none on either axis. Generated-output exclusions: none.

The complete reports and final files-read lists are retained with the Session trace in .thirdshift-review-7l1coz/standards.md and .thirdshift-review-7l1coz/spec.md; summary.json records {"standards":0,"spec":0}.

Unaddressed findings

Standards

None.

Spec

None.

Security

None.

Security review outcome

No unaddressed introduced findings.

Merge Danger

Door: two-way

A reversible in-process ownership refactor; no data migration or new execution path.

Blast Radius: Cron

Cron jobs metadata, Launch directory discovery and Console Run now choices share the interpreter. Filesystem checks, supported-Pass matching and request/submission validation remain with their existing owners.

Built with codex codex-cli 0.167.1 · gpt-6.1-sol · xhigh