feat(lenses): Exodus 1-10 lens content, batch 1 of Exodus pilot (#820, #1781)

[CraigBuckmaster]

Summary

Opens the Exodus pilot. Same curated-sparse cadence proven across the Genesis pilot, now applied to Exodus 1–10. 41 entries across 10 chapters, all scoring 100/100 on lens_quality_scorer after two minor restatements.

Per-chapter lens distribution

Chapter	Lenses	Notes
ex1 — Israel multiplies; midwives' courage	4	literary, redemptive, canonical, devotional
ex2 — Moses' birth and rescue	5	grammatical, literary, typological, redemptive, devotional
ex3 — Burning bush; "I AM"	6	theology peak — divine name, presence, commission
ex4 — Objections; signs; circumcision crisis	4	grammatical, redemptive, canonical, devotional
ex5 — First confrontation; bricks without straw	3	literary, redemptive, devotional — sparse, frustration
ex6 — Renewed commission; YHWH name	5	grammatical, redemptive, canonical, christocentric, devotional
ex7 — Aaron's staff; first plague	4	literary, typological, redemptive, canonical
ex8 — Plagues 2–4	3	literary, canonical, devotional — sparse, repetitive cycle
ex9 — Plagues 5–7 (incl. Rom 9:17 anchor)	4	redemptive, mission, canonical, devotional
ex10 — Plagues 8–9	3	literary, redemptive, canonical

Total: 41 entries. Average 4.1 lenses/chapter — within the curated-sparse band. ex3 maxed at 6 because it's the theological centre of the book; ex5/ex8/ex10 sparse because they are narrative repetition or frustration chapters where forcing a sixth lens would dilute.

First missional lens of the pilot: ex9. Verse 16 ("that my name may be proclaimed in all the earth") is the OT's most explicit missional charter and is cited at Rom 9:17 — earning the lens a place where Genesis only had it once (gen10/gen12).

Theological highlights

• ex3 christocentric: John 8:58's "before Abraham was, I AM" explicitly cites Ex 3:14 — and the bush that burns without being consumed (v 2) was read by Patristic readers as a type of the incarnation (presence sustaining its vessel without destroying it).
• ex2 typological: Moses saved from genocide, drawn from water, called out of Egypt — the pattern Matthew 2:13–15 explicitly applies to Christ saved from Herod and called out of Egypt. Anchored, not invented.
• ex9 mission: v 16's "that my name may be proclaimed in all the earth" → Rom 9:17 (Paul's argument from Pharaoh in his doctrine of election). Exodus's local conflict becomes the New Testament's missional theology.
• ex10 canonical: Three days of darkness (vv 22–23) anticipates Amos 8:9, Joel 2:31, and the three-hour darkness at the cross (Mt 27:45) — judgement preceding redemption is the canonical pattern.
• ex7 typological: Aaron's staff "swallows" (Hebrew בלע, bala') the magicians' (v 12) — same verb at 1 Cor 15:54 ("death swallowed up in victory"). Genuine Hebrew wordplay running across testaments.
• ex2 literary: Moses' "basket" in v 3 is Hebrew tebah — used only of Noah's ark elsewhere (Gen 6:14). The narrative literally floats salvation on the same word twice.

Pipeline gates (all green)

Gate	Result
schema_validator.py	141484 passed / 0 failed / 19 unrelated content warnings
lens_quality_scorer.py	41/41 entries at 100/100 (all four dimensions full marks)
build_sqlite.py	8 lenses, 209 contents
validate_sqlite.py	101 passed / 0 failed

Drafting discipline — lessons applied and surfaced

The 250-char ceiling held — no entry exceeded 245 chars on first draft.

Two restatements needed, both for relevance (themes-panel word overlap above the 60% threshold):

Entry	First-draft overlap	After restate
ex2 redemptive (v 24, "God remembered his covenant")	76%	<50%
ex6 grammatical (v 3, YHWH self-disclosure)	61%	<50%

Both were 93/100 on the scorer (passing the 90 floor as drafted) but lifted to 100 by switching to less-paraphrased phrasing — ex2's restatement now anchors on Hebrew zakar and parallels at Gen 8:1 / 1 Sam 1:19; ex6's now leans on yada' and the relational/informational distinction.

Pattern to flag for future Exodus batches: chapters with iconic single verses where the themes panel itself is a dense paraphrase of that verse will hit the plagiarism threshold easily. Worth building this into the drafting checklist:

• For verse anchors that are quoted in the themes panel (ex 2:24, 3:14, 6:3, 9:16, etc.), draft the entry deliberately to anchor on a different lexical handle than the themes panel uses.
• A Hebrew/Greek term, a canonical parallel, or a different verse in the same chapter all work as alternative handles.

No banned filler patterns; no should_avoid matches; no should_avoid lens-specific avoid-phrases. The lens quality cadence is healthy.

Watch list for tier-2 audit

Six entries flagged for editor scrutiny on a deeper read.

1. ex3 typological — "Patristic readers saw a type of the incarnation" (bush-not-consumed). Defensible (Gregory of Nyssa, Cyril of Alexandria, the standard Eastern reading) but brings patristic-specific imagery; some Reformed editors might prefer "anticipation" framing.
2. ex3 christocentric — Tying Jn 8:58 directly to Ex 3:14 is the standard Christological reading but compresses fast; an editor may want to note the linguistic parallel (egō eimi / ehyeh) more explicitly.
3. ex7 typological — bala' ("swallow") connecting Aaron's staff to 1 Cor 15:54 is real lexical data but is a pattern claim that some traditions would prefer as "echoes" rather than a hard typology.
4. ex9 mission — quoting v 16 then citing Rom 9:17 within a four-line entry: tightly compressed; sound but worth pacing slower in a tier-2 expansion.
5. ex10 canonical — chaining vv 22–23 → Amos 8:9 → Joel 2:31 → Mt 27:45 in one sentence is theologically defensible but covers a lot of ground; an editor may want to slow this connection.
6. ex2 typological — Mt 2:13–15 is an explicit citation; this one is on solid ground, but the framing "the pattern Matthew 2:13–15 cites" reads slightly forced. Cosmetic.

Branch / merge notes

• Branched off master at b183f7b (pre-feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796).
• PR feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 (Genesis batch 5, gen 41–50) is still in review. This PR's chapter files are non-overlapping with feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796's — Exodus directory is entirely separate from Genesis. Merge order is flexible.
• Reported lens_contents row count of 209 assumes feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 has not landed. After both PRs land, total will be 255 (Genesis 214 + Exodus 41).

Rollback plan

Revert the merge commit. The 10 new chapter files are additive and isolated under content/hermeneutic_lenses/chapters/; no existing content, schema, or app code is touched. SQLite rebuilds deterministically.

Roadmap

After this lands, batches 2–4 of Exodus complete the Exodus pilot:

• Batch 2: ex11–ex20 (Passover, Red Sea, song of Moses, manna, water from rock — major typological density expected, possibly 5+ lens average)
• Batch 3: ex21–ex30 (Book of the Covenant, tabernacle instructions — mostly legal/architectural, lower lens average)
• Batch 4: ex31–ex40 (Sabbath, golden calf, glory, tabernacle construction — golden calf chapter likely 6 lenses)

Refs #820. Closes part of #1781.

[github-actions]

Content Pipeline Results

✅ All pipeline checks passed

Step	Status	Details
Schema Validation	✅	142240 passed, 0 failed
Build DB	✅	—
DB Integrity	✅	101 passed, 0 failed

[github-actions]

Test Results

✅ All tests passed

	Passed	Failed	Total
Tests	✅ 4001	❌ 0	4001
Suites	✅ 532	❌ 0	532

Coverage

Statements	Branches	Functions	Lines
—	—	—	—

⏱️ Duration: 93.0s

[CraigBuckmaster]

Summary

Opens the Exodus pilot. Same curated-sparse cadence proven across the Genesis pilot, now applied to Exodus 1–10. 41 entries across 10 chapters, all scoring 100/100 on lens_quality_scorer after two minor restatements.

Per-chapter lens distribution

Chapter	Lenses	Notes
ex1 — Israel multiplies; midwives' courage	4	literary, redemptive, canonical, devotional
ex2 — Moses' birth and rescue	5	grammatical, literary, typological, redemptive, devotional
ex3 — Burning bush; "I AM"	6	theology peak — divine name, presence, commission
ex4 — Objections; signs; circumcision crisis	4	grammatical, redemptive, canonical, devotional
ex5 — First confrontation; bricks without straw	3	literary, redemptive, devotional — sparse, frustration
ex6 — Renewed commission; YHWH name	5	grammatical, redemptive, canonical, christocentric, devotional
ex7 — Aaron's staff; first plague	4	literary, typological, redemptive, canonical
ex8 — Plagues 2–4	3	literary, canonical, devotional — sparse, repetitive cycle
ex9 — Plagues 5–7 (incl. Rom 9:17 anchor)	4	redemptive, mission, canonical, devotional
ex10 — Plagues 8–9	3	literary, redemptive, canonical

Total: 41 entries. Average 4.1 lenses/chapter — within the curated-sparse band. ex3 maxed at 6 because it's the theological centre of the book; ex5/ex8/ex10 sparse because they are narrative repetition or frustration chapters where forcing a sixth lens would dilute.

First missional lens of the pilot: ex9. Verse 16 ("that my name may be proclaimed in all the earth") is the OT's most explicit missional charter and is cited at Rom 9:17 — earning the lens a place where Genesis only had it once (gen10/gen12).

Theological highlights

• ex3 christocentric: John 8:58's "before Abraham was, I AM" explicitly cites Ex 3:14 — and the bush that burns without being consumed (v 2) was read by Patristic readers as a type of the incarnation (presence sustaining its vessel without destroying it).
• ex2 typological: Moses saved from genocide, drawn from water, called out of Egypt — the pattern Matthew 2:13–15 explicitly applies to Christ saved from Herod and called out of Egypt. Anchored, not invented.
• ex9 mission: v 16's "that my name may be proclaimed in all the earth" → Rom 9:17 (Paul's argument from Pharaoh in his doctrine of election). Exodus's local conflict becomes the New Testament's missional theology.
• ex10 canonical: Three days of darkness (vv 22–23) anticipates Amos 8:9, Joel 2:31, and the three-hour darkness at the cross (Mt 27:45) — judgement preceding redemption is the canonical pattern.
• ex7 typological: Aaron's staff "swallows" (Hebrew בלע, bala') the magicians' (v 12) — same verb at 1 Cor 15:54 ("death swallowed up in victory"). Genuine Hebrew wordplay running across testaments.
• ex2 literary: Moses' "basket" in v 3 is Hebrew tebah — used only of Noah's ark elsewhere (Gen 6:14). The narrative literally floats salvation on the same word twice.

Pipeline gates (all green)

Gate	Result
schema_validator.py	141484 passed / 0 failed / 19 unrelated content warnings
lens_quality_scorer.py	41/41 entries at 100/100 (all four dimensions full marks)
build_sqlite.py	8 lenses, 209 contents
validate_sqlite.py	101 passed / 0 failed

Drafting discipline — lessons applied and surfaced

The 250-char ceiling held — no entry exceeded 245 chars on first draft.

Two restatements needed, both for relevance (themes-panel word overlap above the 60% threshold):

Entry	First-draft overlap	After restate
ex2 redemptive (v 24, "God remembered his covenant")	76%	<50%
ex6 grammatical (v 3, YHWH self-disclosure)	61%	<50%

Both were 93/100 on the scorer (passing the 90 floor as drafted) but lifted to 100 by switching to less-paraphrased phrasing — ex2's restatement now anchors on Hebrew zakar and parallels at Gen 8:1 / 1 Sam 1:19; ex6's now leans on yada' and the relational/informational distinction.

Pattern to flag for future Exodus batches: chapters with iconic single verses where the themes panel itself is a dense paraphrase of that verse will hit the plagiarism threshold easily. Worth building this into the drafting checklist:

• For verse anchors that are quoted in the themes panel (ex 2:24, 3:14, 6:3, 9:16, etc.), draft the entry deliberately to anchor on a different lexical handle than the themes panel uses.
• A Hebrew/Greek term, a canonical parallel, or a different verse in the same chapter all work as alternative handles.

No banned filler patterns; no should_avoid matches; no should_avoid lens-specific avoid-phrases. The lens quality cadence is healthy.

Watch list for tier-2 audit

Six entries flagged for editor scrutiny on a deeper read.

1. ex3 typological — "Patristic readers saw a type of the incarnation" (bush-not-consumed). Defensible (Gregory of Nyssa, Cyril of Alexandria, the standard Eastern reading) but brings patristic-specific imagery; some Reformed editors might prefer "anticipation" framing.
2. ex3 christocentric — Tying Jn 8:58 directly to Ex 3:14 is the standard Christological reading but compresses fast; an editor may want to note the linguistic parallel (egō eimi / ehyeh) more explicitly.
3. ex7 typological — bala' ("swallow") connecting Aaron's staff to 1 Cor 15:54 is real lexical data but is a pattern claim that some traditions would prefer as "echoes" rather than a hard typology.
4. ex9 mission — quoting v 16 then citing Rom 9:17 within a four-line entry: tightly compressed; sound but worth pacing slower in a tier-2 expansion.
5. ex10 canonical — chaining vv 22–23 → Amos 8:9 → Joel 2:31 → Mt 27:45 in one sentence is theologically defensible but covers a lot of ground; an editor may want to slow this connection.
6. ex2 typological — Mt 2:13–15 is an explicit citation; this one is on solid ground, but the framing "the pattern Matthew 2:13–15 cites" reads slightly forced. Cosmetic.

Branch / merge notes

• Branched off master at b183f7b (pre-feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796).
• PR feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 (Genesis batch 5, gen 41–50) is still in review. This PR's chapter files are non-overlapping with feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796's — Exodus directory is entirely separate from Genesis. Merge order is flexible.
• Reported lens_contents row count of 209 assumes feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 has not landed. After both PRs land, total will be 255 (Genesis 214 + Exodus 41).

Rollback plan

Revert the merge commit. The 10 new chapter files are additive and isolated under content/hermeneutic_lenses/chapters/; no existing content, schema, or app code is touched. SQLite rebuilds deterministically.

Roadmap

After this lands, batches 2–4 of Exodus complete the Exodus pilot:

• Batch 2: ex11–ex20 (Passover, Red Sea, song of Moses, manna, water from rock — major typological density expected, possibly 5+ lens average)
• Batch 3: ex21–ex30 (Book of the Covenant, tabernacle instructions — mostly legal/architectural, lower lens average)
• Batch 4: ex31–ex40 (Sabbath, golden calf, glory, tabernacle construction — golden calf chapter likely 6 lenses)

Refs #820. Closes part of #1781.

[github-actions]

Content Pipeline Results

✅ All pipeline checks passed

Step	Status	Details
Schema Validation	✅	142240 passed, 0 failed
Build DB	✅	—
DB Integrity	✅	101 passed, 0 failed

[github-actions]

Test Results

✅ All tests passed

	Passed	Failed	Total
Tests	✅ 4001	❌ 0	4001
Suites	✅ 532	❌ 0	532

Coverage

Statements	Branches	Functions	Lines
—	—	—	—

⏱️ Duration: 93.0s

[CraigBuckmaster]

Summary

Opens the Exodus pilot. Same curated-sparse cadence proven across the Genesis pilot, now applied to Exodus 1–10. 41 entries across 10 chapters, all scoring 100/100 on lens_quality_scorer after two minor restatements.

Per-chapter lens distribution

Chapter	Lenses	Notes
ex1 — Israel multiplies; midwives' courage	4	literary, redemptive, canonical, devotional
ex2 — Moses' birth and rescue	5	grammatical, literary, typological, redemptive, devotional
ex3 — Burning bush; "I AM"	6	theology peak — divine name, presence, commission
ex4 — Objections; signs; circumcision crisis	4	grammatical, redemptive, canonical, devotional
ex5 — First confrontation; bricks without straw	3	literary, redemptive, devotional — sparse, frustration
ex6 — Renewed commission; YHWH name	5	grammatical, redemptive, canonical, christocentric, devotional
ex7 — Aaron's staff; first plague	4	literary, typological, redemptive, canonical
ex8 — Plagues 2–4	3	literary, canonical, devotional — sparse, repetitive cycle
ex9 — Plagues 5–7 (incl. Rom 9:17 anchor)	4	redemptive, mission, canonical, devotional
ex10 — Plagues 8–9	3	literary, redemptive, canonical

Total: 41 entries. Average 4.1 lenses/chapter — within the curated-sparse band. ex3 maxed at 6 because it's the theological centre of the book; ex5/ex8/ex10 sparse because they are narrative repetition or frustration chapters where forcing a sixth lens would dilute.

First missional lens of the pilot: ex9. Verse 16 ("that my name may be proclaimed in all the earth") is the OT's most explicit missional charter and is cited at Rom 9:17 — earning the lens a place where Genesis only had it once (gen10/gen12).

Theological highlights

• ex3 christocentric: John 8:58's "before Abraham was, I AM" explicitly cites Ex 3:14 — and the bush that burns without being consumed (v 2) was read by Patristic readers as a type of the incarnation (presence sustaining its vessel without destroying it).
• ex2 typological: Moses saved from genocide, drawn from water, called out of Egypt — the pattern Matthew 2:13–15 explicitly applies to Christ saved from Herod and called out of Egypt. Anchored, not invented.
• ex9 mission: v 16's "that my name may be proclaimed in all the earth" → Rom 9:17 (Paul's argument from Pharaoh in his doctrine of election). Exodus's local conflict becomes the New Testament's missional theology.
• ex10 canonical: Three days of darkness (vv 22–23) anticipates Amos 8:9, Joel 2:31, and the three-hour darkness at the cross (Mt 27:45) — judgement preceding redemption is the canonical pattern.
• ex7 typological: Aaron's staff "swallows" (Hebrew בלע, bala') the magicians' (v 12) — same verb at 1 Cor 15:54 ("death swallowed up in victory"). Genuine Hebrew wordplay running across testaments.
• ex2 literary: Moses' "basket" in v 3 is Hebrew tebah — used only of Noah's ark elsewhere (Gen 6:14). The narrative literally floats salvation on the same word twice.

Pipeline gates (all green)

Gate	Result
schema_validator.py	141484 passed / 0 failed / 19 unrelated content warnings
lens_quality_scorer.py	41/41 entries at 100/100 (all four dimensions full marks)
build_sqlite.py	8 lenses, 209 contents
validate_sqlite.py	101 passed / 0 failed

Drafting discipline — lessons applied and surfaced

The 250-char ceiling held — no entry exceeded 245 chars on first draft.

Two restatements needed, both for relevance (themes-panel word overlap above the 60% threshold):

Entry	First-draft overlap	After restate
ex2 redemptive (v 24, "God remembered his covenant")	76%	<50%
ex6 grammatical (v 3, YHWH self-disclosure)	61%	<50%

Both were 93/100 on the scorer (passing the 90 floor as drafted) but lifted to 100 by switching to less-paraphrased phrasing — ex2's restatement now anchors on Hebrew zakar and parallels at Gen 8:1 / 1 Sam 1:19; ex6's now leans on yada' and the relational/informational distinction.

Pattern to flag for future Exodus batches: chapters with iconic single verses where the themes panel itself is a dense paraphrase of that verse will hit the plagiarism threshold easily. Worth building this into the drafting checklist:

• For verse anchors that are quoted in the themes panel (ex 2:24, 3:14, 6:3, 9:16, etc.), draft the entry deliberately to anchor on a different lexical handle than the themes panel uses.
• A Hebrew/Greek term, a canonical parallel, or a different verse in the same chapter all work as alternative handles.

No banned filler patterns; no should_avoid matches; no should_avoid lens-specific avoid-phrases. The lens quality cadence is healthy.

Watch list for tier-2 audit

Six entries flagged for editor scrutiny on a deeper read.

1. ex3 typological — "Patristic readers saw a type of the incarnation" (bush-not-consumed). Defensible (Gregory of Nyssa, Cyril of Alexandria, the standard Eastern reading) but brings patristic-specific imagery; some Reformed editors might prefer "anticipation" framing.
2. ex3 christocentric — Tying Jn 8:58 directly to Ex 3:14 is the standard Christological reading but compresses fast; an editor may want to note the linguistic parallel (egō eimi / ehyeh) more explicitly.
3. ex7 typological — bala' ("swallow") connecting Aaron's staff to 1 Cor 15:54 is real lexical data but is a pattern claim that some traditions would prefer as "echoes" rather than a hard typology.
4. ex9 mission — quoting v 16 then citing Rom 9:17 within a four-line entry: tightly compressed; sound but worth pacing slower in a tier-2 expansion.
5. ex10 canonical — chaining vv 22–23 → Amos 8:9 → Joel 2:31 → Mt 27:45 in one sentence is theologically defensible but covers a lot of ground; an editor may want to slow this connection.
6. ex2 typological — Mt 2:13–15 is an explicit citation; this one is on solid ground, but the framing "the pattern Matthew 2:13–15 cites" reads slightly forced. Cosmetic.

Branch / merge notes

• Branched off master at b183f7b (pre-feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796).
• PR feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 (Genesis batch 5, gen 41–50) is still in review. This PR's chapter files are non-overlapping with feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796's — Exodus directory is entirely separate from Genesis. Merge order is flexible.
• Reported lens_contents row count of 209 assumes feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 has not landed. After both PRs land, total will be 255 (Genesis 214 + Exodus 41).

Rollback plan

Revert the merge commit. The 10 new chapter files are additive and isolated under content/hermeneutic_lenses/chapters/; no existing content, schema, or app code is touched. SQLite rebuilds deterministically.

Roadmap

After this lands, batches 2–4 of Exodus complete the Exodus pilot:

• Batch 2: ex11–ex20 (Passover, Red Sea, song of Moses, manna, water from rock — major typological density expected, possibly 5+ lens average)
• Batch 3: ex21–ex30 (Book of the Covenant, tabernacle instructions — mostly legal/architectural, lower lens average)
• Batch 4: ex31–ex40 (Sabbath, golden calf, glory, tabernacle construction — golden calf chapter likely 6 lenses)

Refs #820. Closes part of #1781.

[github-actions]

Content Pipeline Results

✅ All pipeline checks passed

Step	Status	Details
Schema Validation	✅	142240 passed, 0 failed
Build DB	✅	—
DB Integrity	✅	101 passed, 0 failed

[github-actions]

Test Results

✅ All tests passed

	Passed	Failed	Total
Tests	✅ 4001	❌ 0	4001
Suites	✅ 532	❌ 0	532

Coverage

Statements	Branches	Functions	Lines
—	—	—	—

⏱️ Duration: 93.0s

[CraigBuckmaster]

Summary

Opens the Exodus pilot. Same curated-sparse cadence proven across the Genesis pilot, now applied to Exodus 1–10. 41 entries across 10 chapters, all scoring 100/100 on lens_quality_scorer after two minor restatements.

Per-chapter lens distribution

Chapter	Lenses	Notes
ex1 — Israel multiplies; midwives' courage	4	literary, redemptive, canonical, devotional
ex2 — Moses' birth and rescue	5	grammatical, literary, typological, redemptive, devotional
ex3 — Burning bush; "I AM"	6	theology peak — divine name, presence, commission
ex4 — Objections; signs; circumcision crisis	4	grammatical, redemptive, canonical, devotional
ex5 — First confrontation; bricks without straw	3	literary, redemptive, devotional — sparse, frustration
ex6 — Renewed commission; YHWH name	5	grammatical, redemptive, canonical, christocentric, devotional
ex7 — Aaron's staff; first plague	4	literary, typological, redemptive, canonical
ex8 — Plagues 2–4	3	literary, canonical, devotional — sparse, repetitive cycle
ex9 — Plagues 5–7 (incl. Rom 9:17 anchor)	4	redemptive, mission, canonical, devotional
ex10 — Plagues 8–9	3	literary, redemptive, canonical

Total: 41 entries. Average 4.1 lenses/chapter — within the curated-sparse band. ex3 maxed at 6 because it's the theological centre of the book; ex5/ex8/ex10 sparse because they are narrative repetition or frustration chapters where forcing a sixth lens would dilute.

First missional lens of the pilot: ex9. Verse 16 ("that my name may be proclaimed in all the earth") is the OT's most explicit missional charter and is cited at Rom 9:17 — earning the lens a place where Genesis only had it once (gen10/gen12).

Theological highlights

• ex3 christocentric: John 8:58's "before Abraham was, I AM" explicitly cites Ex 3:14 — and the bush that burns without being consumed (v 2) was read by Patristic readers as a type of the incarnation (presence sustaining its vessel without destroying it).
• ex2 typological: Moses saved from genocide, drawn from water, called out of Egypt — the pattern Matthew 2:13–15 explicitly applies to Christ saved from Herod and called out of Egypt. Anchored, not invented.
• ex9 mission: v 16's "that my name may be proclaimed in all the earth" → Rom 9:17 (Paul's argument from Pharaoh in his doctrine of election). Exodus's local conflict becomes the New Testament's missional theology.
• ex10 canonical: Three days of darkness (vv 22–23) anticipates Amos 8:9, Joel 2:31, and the three-hour darkness at the cross (Mt 27:45) — judgement preceding redemption is the canonical pattern.
• ex7 typological: Aaron's staff "swallows" (Hebrew בלע, bala') the magicians' (v 12) — same verb at 1 Cor 15:54 ("death swallowed up in victory"). Genuine Hebrew wordplay running across testaments.
• ex2 literary: Moses' "basket" in v 3 is Hebrew tebah — used only of Noah's ark elsewhere (Gen 6:14). The narrative literally floats salvation on the same word twice.

Pipeline gates (all green)

Gate	Result
schema_validator.py	141484 passed / 0 failed / 19 unrelated content warnings
lens_quality_scorer.py	41/41 entries at 100/100 (all four dimensions full marks)
build_sqlite.py	8 lenses, 209 contents
validate_sqlite.py	101 passed / 0 failed

Drafting discipline — lessons applied and surfaced

The 250-char ceiling held — no entry exceeded 245 chars on first draft.

Two restatements needed, both for relevance (themes-panel word overlap above the 60% threshold):

Entry	First-draft overlap	After restate
ex2 redemptive (v 24, "God remembered his covenant")	76%	<50%
ex6 grammatical (v 3, YHWH self-disclosure)	61%	<50%

Both were 93/100 on the scorer (passing the 90 floor as drafted) but lifted to 100 by switching to less-paraphrased phrasing — ex2's restatement now anchors on Hebrew zakar and parallels at Gen 8:1 / 1 Sam 1:19; ex6's now leans on yada' and the relational/informational distinction.

Pattern to flag for future Exodus batches: chapters with iconic single verses where the themes panel itself is a dense paraphrase of that verse will hit the plagiarism threshold easily. Worth building this into the drafting checklist:

• For verse anchors that are quoted in the themes panel (ex 2:24, 3:14, 6:3, 9:16, etc.), draft the entry deliberately to anchor on a different lexical handle than the themes panel uses.
• A Hebrew/Greek term, a canonical parallel, or a different verse in the same chapter all work as alternative handles.

No banned filler patterns; no should_avoid matches; no should_avoid lens-specific avoid-phrases. The lens quality cadence is healthy.

Watch list for tier-2 audit

Six entries flagged for editor scrutiny on a deeper read.

1. ex3 typological — "Patristic readers saw a type of the incarnation" (bush-not-consumed). Defensible (Gregory of Nyssa, Cyril of Alexandria, the standard Eastern reading) but brings patristic-specific imagery; some Reformed editors might prefer "anticipation" framing.
2. ex3 christocentric — Tying Jn 8:58 directly to Ex 3:14 is the standard Christological reading but compresses fast; an editor may want to note the linguistic parallel (egō eimi / ehyeh) more explicitly.
3. ex7 typological — bala' ("swallow") connecting Aaron's staff to 1 Cor 15:54 is real lexical data but is a pattern claim that some traditions would prefer as "echoes" rather than a hard typology.
4. ex9 mission — quoting v 16 then citing Rom 9:17 within a four-line entry: tightly compressed; sound but worth pacing slower in a tier-2 expansion.
5. ex10 canonical — chaining vv 22–23 → Amos 8:9 → Joel 2:31 → Mt 27:45 in one sentence is theologically defensible but covers a lot of ground; an editor may want to slow this connection.
6. ex2 typological — Mt 2:13–15 is an explicit citation; this one is on solid ground, but the framing "the pattern Matthew 2:13–15 cites" reads slightly forced. Cosmetic.

Branch / merge notes

• Branched off master at b183f7b (pre-feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796).
• PR feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 (Genesis batch 5, gen 41–50) is still in review. This PR's chapter files are non-overlapping with feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796's — Exodus directory is entirely separate from Genesis. Merge order is flexible.
• Reported lens_contents row count of 209 assumes feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 has not landed. After both PRs land, total will be 255 (Genesis 214 + Exodus 41).

Rollback plan

Revert the merge commit. The 10 new chapter files are additive and isolated under content/hermeneutic_lenses/chapters/; no existing content, schema, or app code is touched. SQLite rebuilds deterministically.

Roadmap

After this lands, batches 2–4 of Exodus complete the Exodus pilot:

• Batch 2: ex11–ex20 (Passover, Red Sea, song of Moses, manna, water from rock — major typological density expected, possibly 5+ lens average)
• Batch 3: ex21–ex30 (Book of the Covenant, tabernacle instructions — mostly legal/architectural, lower lens average)
• Batch 4: ex31–ex40 (Sabbath, golden calf, glory, tabernacle construction — golden calf chapter likely 6 lenses)

Refs #820. Closes part of #1781.

[github-actions]

Content Pipeline Results

✅ All pipeline checks passed

Step	Status	Details
Schema Validation	✅	142240 passed, 0 failed
Build DB	✅	—
DB Integrity	✅	101 passed, 0 failed

[github-actions]

Test Results

✅ All tests passed

	Passed	Failed	Total
Tests	✅ 4001	❌ 0	4001
Suites	✅ 532	❌ 0	532

Coverage

Statements	Branches	Functions	Lines
—	—	—	—

⏱️ Duration: 93.0s

[CraigBuckmaster]

Summary

Opens the Exodus pilot. Same curated-sparse cadence proven across the Genesis pilot, now applied to Exodus 1–10. 41 entries across 10 chapters, all scoring 100/100 on lens_quality_scorer after two minor restatements.

Per-chapter lens distribution

Chapter	Lenses	Notes
ex1 — Israel multiplies; midwives' courage	4	literary, redemptive, canonical, devotional
ex2 — Moses' birth and rescue	5	grammatical, literary, typological, redemptive, devotional
ex3 — Burning bush; "I AM"	6	theology peak — divine name, presence, commission
ex4 — Objections; signs; circumcision crisis	4	grammatical, redemptive, canonical, devotional
ex5 — First confrontation; bricks without straw	3	literary, redemptive, devotional — sparse, frustration
ex6 — Renewed commission; YHWH name	5	grammatical, redemptive, canonical, christocentric, devotional
ex7 — Aaron's staff; first plague	4	literary, typological, redemptive, canonical
ex8 — Plagues 2–4	3	literary, canonical, devotional — sparse, repetitive cycle
ex9 — Plagues 5–7 (incl. Rom 9:17 anchor)	4	redemptive, mission, canonical, devotional
ex10 — Plagues 8–9	3	literary, redemptive, canonical

Total: 41 entries. Average 4.1 lenses/chapter — within the curated-sparse band. ex3 maxed at 6 because it's the theological centre of the book; ex5/ex8/ex10 sparse because they are narrative repetition or frustration chapters where forcing a sixth lens would dilute.

First missional lens of the pilot: ex9. Verse 16 ("that my name may be proclaimed in all the earth") is the OT's most explicit missional charter and is cited at Rom 9:17 — earning the lens a place where Genesis only had it once (gen10/gen12).

Theological highlights

• ex3 christocentric: John 8:58's "before Abraham was, I AM" explicitly cites Ex 3:14 — and the bush that burns without being consumed (v 2) was read by Patristic readers as a type of the incarnation (presence sustaining its vessel without destroying it).
• ex2 typological: Moses saved from genocide, drawn from water, called out of Egypt — the pattern Matthew 2:13–15 explicitly applies to Christ saved from Herod and called out of Egypt. Anchored, not invented.
• ex9 mission: v 16's "that my name may be proclaimed in all the earth" → Rom 9:17 (Paul's argument from Pharaoh in his doctrine of election). Exodus's local conflict becomes the New Testament's missional theology.
• ex10 canonical: Three days of darkness (vv 22–23) anticipates Amos 8:9, Joel 2:31, and the three-hour darkness at the cross (Mt 27:45) — judgement preceding redemption is the canonical pattern.
• ex7 typological: Aaron's staff "swallows" (Hebrew בלע, bala') the magicians' (v 12) — same verb at 1 Cor 15:54 ("death swallowed up in victory"). Genuine Hebrew wordplay running across testaments.
• ex2 literary: Moses' "basket" in v 3 is Hebrew tebah — used only of Noah's ark elsewhere (Gen 6:14). The narrative literally floats salvation on the same word twice.

Pipeline gates (all green)

Gate	Result
schema_validator.py	141484 passed / 0 failed / 19 unrelated content warnings
lens_quality_scorer.py	41/41 entries at 100/100 (all four dimensions full marks)
build_sqlite.py	8 lenses, 209 contents
validate_sqlite.py	101 passed / 0 failed

Drafting discipline — lessons applied and surfaced

The 250-char ceiling held — no entry exceeded 245 chars on first draft.

Two restatements needed, both for relevance (themes-panel word overlap above the 60% threshold):

Entry	First-draft overlap	After restate
ex2 redemptive (v 24, "God remembered his covenant")	76%	<50%
ex6 grammatical (v 3, YHWH self-disclosure)	61%	<50%

Both were 93/100 on the scorer (passing the 90 floor as drafted) but lifted to 100 by switching to less-paraphrased phrasing — ex2's restatement now anchors on Hebrew zakar and parallels at Gen 8:1 / 1 Sam 1:19; ex6's now leans on yada' and the relational/informational distinction.

Pattern to flag for future Exodus batches: chapters with iconic single verses where the themes panel itself is a dense paraphrase of that verse will hit the plagiarism threshold easily. Worth building this into the drafting checklist:

• For verse anchors that are quoted in the themes panel (ex 2:24, 3:14, 6:3, 9:16, etc.), draft the entry deliberately to anchor on a different lexical handle than the themes panel uses.
• A Hebrew/Greek term, a canonical parallel, or a different verse in the same chapter all work as alternative handles.

No banned filler patterns; no should_avoid matches; no should_avoid lens-specific avoid-phrases. The lens quality cadence is healthy.

Watch list for tier-2 audit

Six entries flagged for editor scrutiny on a deeper read.

1. ex3 typological — "Patristic readers saw a type of the incarnation" (bush-not-consumed). Defensible (Gregory of Nyssa, Cyril of Alexandria, the standard Eastern reading) but brings patristic-specific imagery; some Reformed editors might prefer "anticipation" framing.
2. ex3 christocentric — Tying Jn 8:58 directly to Ex 3:14 is the standard Christological reading but compresses fast; an editor may want to note the linguistic parallel (egō eimi / ehyeh) more explicitly.
3. ex7 typological — bala' ("swallow") connecting Aaron's staff to 1 Cor 15:54 is real lexical data but is a pattern claim that some traditions would prefer as "echoes" rather than a hard typology.
4. ex9 mission — quoting v 16 then citing Rom 9:17 within a four-line entry: tightly compressed; sound but worth pacing slower in a tier-2 expansion.
5. ex10 canonical — chaining vv 22–23 → Amos 8:9 → Joel 2:31 → Mt 27:45 in one sentence is theologically defensible but covers a lot of ground; an editor may want to slow this connection.
6. ex2 typological — Mt 2:13–15 is an explicit citation; this one is on solid ground, but the framing "the pattern Matthew 2:13–15 cites" reads slightly forced. Cosmetic.

Branch / merge notes

• Branched off master at b183f7b (pre-feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796).
• PR feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 (Genesis batch 5, gen 41–50) is still in review. This PR's chapter files are non-overlapping with feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796's — Exodus directory is entirely separate from Genesis. Merge order is flexible.
• Reported lens_contents row count of 209 assumes feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 has not landed. After both PRs land, total will be 255 (Genesis 214 + Exodus 41).

Rollback plan

Revert the merge commit. The 10 new chapter files are additive and isolated under content/hermeneutic_lenses/chapters/; no existing content, schema, or app code is touched. SQLite rebuilds deterministically.

Roadmap

After this lands, batches 2–4 of Exodus complete the Exodus pilot:

• Batch 2: ex11–ex20 (Passover, Red Sea, song of Moses, manna, water from rock — major typological density expected, possibly 5+ lens average)
• Batch 3: ex21–ex30 (Book of the Covenant, tabernacle instructions — mostly legal/architectural, lower lens average)
• Batch 4: ex31–ex40 (Sabbath, golden calf, glory, tabernacle construction — golden calf chapter likely 6 lenses)

Refs #820. Closes part of #1781.

[github-actions]

Content Pipeline Results

✅ All pipeline checks passed

Step	Status	Details
Schema Validation	✅	142240 passed, 0 failed
Build DB	✅	—
DB Integrity	✅	101 passed, 0 failed

[github-actions]

Test Results

✅ All tests passed

	Passed	Failed	Total
Tests	✅ 4001	❌ 0	4001
Suites	✅ 532	❌ 0	532

Coverage

Statements	Branches	Functions	Lines
—	—	—	—

⏱️ Duration: 93.0s

[CraigBuckmaster]

Summary

Opens the Exodus pilot. Same curated-sparse cadence proven across the Genesis pilot, now applied to Exodus 1–10. 41 entries across 10 chapters, all scoring 100/100 on lens_quality_scorer after two minor restatements.

Per-chapter lens distribution

Chapter	Lenses	Notes
ex1 — Israel multiplies; midwives' courage	4	literary, redemptive, canonical, devotional
ex2 — Moses' birth and rescue	5	grammatical, literary, typological, redemptive, devotional
ex3 — Burning bush; "I AM"	6	theology peak — divine name, presence, commission
ex4 — Objections; signs; circumcision crisis	4	grammatical, redemptive, canonical, devotional
ex5 — First confrontation; bricks without straw	3	literary, redemptive, devotional — sparse, frustration
ex6 — Renewed commission; YHWH name	5	grammatical, redemptive, canonical, christocentric, devotional
ex7 — Aaron's staff; first plague	4	literary, typological, redemptive, canonical
ex8 — Plagues 2–4	3	literary, canonical, devotional — sparse, repetitive cycle
ex9 — Plagues 5–7 (incl. Rom 9:17 anchor)	4	redemptive, mission, canonical, devotional
ex10 — Plagues 8–9	3	literary, redemptive, canonical

Total: 41 entries. Average 4.1 lenses/chapter — within the curated-sparse band. ex3 maxed at 6 because it's the theological centre of the book; ex5/ex8/ex10 sparse because they are narrative repetition or frustration chapters where forcing a sixth lens would dilute.

First missional lens of the pilot: ex9. Verse 16 ("that my name may be proclaimed in all the earth") is the OT's most explicit missional charter and is cited at Rom 9:17 — earning the lens a place where Genesis only had it once (gen10/gen12).

Theological highlights

• ex3 christocentric: John 8:58's "before Abraham was, I AM" explicitly cites Ex 3:14 — and the bush that burns without being consumed (v 2) was read by Patristic readers as a type of the incarnation (presence sustaining its vessel without destroying it).
• ex2 typological: Moses saved from genocide, drawn from water, called out of Egypt — the pattern Matthew 2:13–15 explicitly applies to Christ saved from Herod and called out of Egypt. Anchored, not invented.
• ex9 mission: v 16's "that my name may be proclaimed in all the earth" → Rom 9:17 (Paul's argument from Pharaoh in his doctrine of election). Exodus's local conflict becomes the New Testament's missional theology.
• ex10 canonical: Three days of darkness (vv 22–23) anticipates Amos 8:9, Joel 2:31, and the three-hour darkness at the cross (Mt 27:45) — judgement preceding redemption is the canonical pattern.
• ex7 typological: Aaron's staff "swallows" (Hebrew בלע, bala') the magicians' (v 12) — same verb at 1 Cor 15:54 ("death swallowed up in victory"). Genuine Hebrew wordplay running across testaments.
• ex2 literary: Moses' "basket" in v 3 is Hebrew tebah — used only of Noah's ark elsewhere (Gen 6:14). The narrative literally floats salvation on the same word twice.

Pipeline gates (all green)

Gate	Result
schema_validator.py	141484 passed / 0 failed / 19 unrelated content warnings
lens_quality_scorer.py	41/41 entries at 100/100 (all four dimensions full marks)
build_sqlite.py	8 lenses, 209 contents
validate_sqlite.py	101 passed / 0 failed

Drafting discipline — lessons applied and surfaced

The 250-char ceiling held — no entry exceeded 245 chars on first draft.

Two restatements needed, both for relevance (themes-panel word overlap above the 60% threshold):

Entry	First-draft overlap	After restate
ex2 redemptive (v 24, "God remembered his covenant")	76%	<50%
ex6 grammatical (v 3, YHWH self-disclosure)	61%	<50%

Both were 93/100 on the scorer (passing the 90 floor as drafted) but lifted to 100 by switching to less-paraphrased phrasing — ex2's restatement now anchors on Hebrew zakar and parallels at Gen 8:1 / 1 Sam 1:19; ex6's now leans on yada' and the relational/informational distinction.

Pattern to flag for future Exodus batches: chapters with iconic single verses where the themes panel itself is a dense paraphrase of that verse will hit the plagiarism threshold easily. Worth building this into the drafting checklist:

• For verse anchors that are quoted in the themes panel (ex 2:24, 3:14, 6:3, 9:16, etc.), draft the entry deliberately to anchor on a different lexical handle than the themes panel uses.
• A Hebrew/Greek term, a canonical parallel, or a different verse in the same chapter all work as alternative handles.

No banned filler patterns; no should_avoid matches; no should_avoid lens-specific avoid-phrases. The lens quality cadence is healthy.

Watch list for tier-2 audit

Six entries flagged for editor scrutiny on a deeper read.

1. ex3 typological — "Patristic readers saw a type of the incarnation" (bush-not-consumed). Defensible (Gregory of Nyssa, Cyril of Alexandria, the standard Eastern reading) but brings patristic-specific imagery; some Reformed editors might prefer "anticipation" framing.
2. ex3 christocentric — Tying Jn 8:58 directly to Ex 3:14 is the standard Christological reading but compresses fast; an editor may want to note the linguistic parallel (egō eimi / ehyeh) more explicitly.
3. ex7 typological — bala' ("swallow") connecting Aaron's staff to 1 Cor 15:54 is real lexical data but is a pattern claim that some traditions would prefer as "echoes" rather than a hard typology.
4. ex9 mission — quoting v 16 then citing Rom 9:17 within a four-line entry: tightly compressed; sound but worth pacing slower in a tier-2 expansion.
5. ex10 canonical — chaining vv 22–23 → Amos 8:9 → Joel 2:31 → Mt 27:45 in one sentence is theologically defensible but covers a lot of ground; an editor may want to slow this connection.
6. ex2 typological — Mt 2:13–15 is an explicit citation; this one is on solid ground, but the framing "the pattern Matthew 2:13–15 cites" reads slightly forced. Cosmetic.

Branch / merge notes

• Branched off master at b183f7b (pre-feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796).
• PR feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 (Genesis batch 5, gen 41–50) is still in review. This PR's chapter files are non-overlapping with feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796's — Exodus directory is entirely separate from Genesis. Merge order is flexible.
• Reported lens_contents row count of 209 assumes feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 has not landed. After both PRs land, total will be 255 (Genesis 214 + Exodus 41).

Rollback plan

Revert the merge commit. The 10 new chapter files are additive and isolated under content/hermeneutic_lenses/chapters/; no existing content, schema, or app code is touched. SQLite rebuilds deterministically.

Roadmap

After this lands, batches 2–4 of Exodus complete the Exodus pilot:

• Batch 2: ex11–ex20 (Passover, Red Sea, song of Moses, manna, water from rock — major typological density expected, possibly 5+ lens average)
• Batch 3: ex21–ex30 (Book of the Covenant, tabernacle instructions — mostly legal/architectural, lower lens average)
• Batch 4: ex31–ex40 (Sabbath, golden calf, glory, tabernacle construction — golden calf chapter likely 6 lenses)

Refs #820. Closes part of #1781.

[github-actions]

Content Pipeline Results

✅ All pipeline checks passed

Step	Status	Details
Schema Validation	✅	142240 passed, 0 failed
Build DB	✅	—
DB Integrity	✅	101 passed, 0 failed

[github-actions]

Test Results

✅ All tests passed

	Passed	Failed	Total
Tests	✅ 4001	❌ 0	4001
Suites	✅ 532	❌ 0	532

Coverage

Statements	Branches	Functions	Lines
—	—	—	—

⏱️ Duration: 93.0s

[CraigBuckmaster]

Summary

Opens the Exodus pilot. Same curated-sparse cadence proven across the Genesis pilot, now applied to Exodus 1–10. 41 entries across 10 chapters, all scoring 100/100 on lens_quality_scorer after two minor restatements.

Per-chapter lens distribution

Chapter	Lenses	Notes
ex1 — Israel multiplies; midwives' courage	4	literary, redemptive, canonical, devotional
ex2 — Moses' birth and rescue	5	grammatical, literary, typological, redemptive, devotional
ex3 — Burning bush; "I AM"	6	theology peak — divine name, presence, commission
ex4 — Objections; signs; circumcision crisis	4	grammatical, redemptive, canonical, devotional
ex5 — First confrontation; bricks without straw	3	literary, redemptive, devotional — sparse, frustration
ex6 — Renewed commission; YHWH name	5	grammatical, redemptive, canonical, christocentric, devotional
ex7 — Aaron's staff; first plague	4	literary, typological, redemptive, canonical
ex8 — Plagues 2–4	3	literary, canonical, devotional — sparse, repetitive cycle
ex9 — Plagues 5–7 (incl. Rom 9:17 anchor)	4	redemptive, mission, canonical, devotional
ex10 — Plagues 8–9	3	literary, redemptive, canonical

Total: 41 entries. Average 4.1 lenses/chapter — within the curated-sparse band. ex3 maxed at 6 because it's the theological centre of the book; ex5/ex8/ex10 sparse because they are narrative repetition or frustration chapters where forcing a sixth lens would dilute.

First missional lens of the pilot: ex9. Verse 16 ("that my name may be proclaimed in all the earth") is the OT's most explicit missional charter and is cited at Rom 9:17 — earning the lens a place where Genesis only had it once (gen10/gen12).

Theological highlights

• ex3 christocentric: John 8:58's "before Abraham was, I AM" explicitly cites Ex 3:14 — and the bush that burns without being consumed (v 2) was read by Patristic readers as a type of the incarnation (presence sustaining its vessel without destroying it).
• ex2 typological: Moses saved from genocide, drawn from water, called out of Egypt — the pattern Matthew 2:13–15 explicitly applies to Christ saved from Herod and called out of Egypt. Anchored, not invented.
• ex9 mission: v 16's "that my name may be proclaimed in all the earth" → Rom 9:17 (Paul's argument from Pharaoh in his doctrine of election). Exodus's local conflict becomes the New Testament's missional theology.
• ex10 canonical: Three days of darkness (vv 22–23) anticipates Amos 8:9, Joel 2:31, and the three-hour darkness at the cross (Mt 27:45) — judgement preceding redemption is the canonical pattern.
• ex7 typological: Aaron's staff "swallows" (Hebrew בלע, bala') the magicians' (v 12) — same verb at 1 Cor 15:54 ("death swallowed up in victory"). Genuine Hebrew wordplay running across testaments.
• ex2 literary: Moses' "basket" in v 3 is Hebrew tebah — used only of Noah's ark elsewhere (Gen 6:14). The narrative literally floats salvation on the same word twice.

Pipeline gates (all green)

Gate	Result
schema_validator.py	141484 passed / 0 failed / 19 unrelated content warnings
lens_quality_scorer.py	41/41 entries at 100/100 (all four dimensions full marks)
build_sqlite.py	8 lenses, 209 contents
validate_sqlite.py	101 passed / 0 failed

Drafting discipline — lessons applied and surfaced

The 250-char ceiling held — no entry exceeded 245 chars on first draft.

Two restatements needed, both for relevance (themes-panel word overlap above the 60% threshold):

Entry	First-draft overlap	After restate
ex2 redemptive (v 24, "God remembered his covenant")	76%	<50%
ex6 grammatical (v 3, YHWH self-disclosure)	61%	<50%

Both were 93/100 on the scorer (passing the 90 floor as drafted) but lifted to 100 by switching to less-paraphrased phrasing — ex2's restatement now anchors on Hebrew zakar and parallels at Gen 8:1 / 1 Sam 1:19; ex6's now leans on yada' and the relational/informational distinction.

Pattern to flag for future Exodus batches: chapters with iconic single verses where the themes panel itself is a dense paraphrase of that verse will hit the plagiarism threshold easily. Worth building this into the drafting checklist:

• For verse anchors that are quoted in the themes panel (ex 2:24, 3:14, 6:3, 9:16, etc.), draft the entry deliberately to anchor on a different lexical handle than the themes panel uses.
• A Hebrew/Greek term, a canonical parallel, or a different verse in the same chapter all work as alternative handles.

No banned filler patterns; no should_avoid matches; no should_avoid lens-specific avoid-phrases. The lens quality cadence is healthy.

Watch list for tier-2 audit

Six entries flagged for editor scrutiny on a deeper read.

1. ex3 typological — "Patristic readers saw a type of the incarnation" (bush-not-consumed). Defensible (Gregory of Nyssa, Cyril of Alexandria, the standard Eastern reading) but brings patristic-specific imagery; some Reformed editors might prefer "anticipation" framing.
2. ex3 christocentric — Tying Jn 8:58 directly to Ex 3:14 is the standard Christological reading but compresses fast; an editor may want to note the linguistic parallel (egō eimi / ehyeh) more explicitly.
3. ex7 typological — bala' ("swallow") connecting Aaron's staff to 1 Cor 15:54 is real lexical data but is a pattern claim that some traditions would prefer as "echoes" rather than a hard typology.
4. ex9 mission — quoting v 16 then citing Rom 9:17 within a four-line entry: tightly compressed; sound but worth pacing slower in a tier-2 expansion.
5. ex10 canonical — chaining vv 22–23 → Amos 8:9 → Joel 2:31 → Mt 27:45 in one sentence is theologically defensible but covers a lot of ground; an editor may want to slow this connection.
6. ex2 typological — Mt 2:13–15 is an explicit citation; this one is on solid ground, but the framing "the pattern Matthew 2:13–15 cites" reads slightly forced. Cosmetic.

Branch / merge notes

• Branched off master at b183f7b (pre-feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796).
• PR feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 (Genesis batch 5, gen 41–50) is still in review. This PR's chapter files are non-overlapping with feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796's — Exodus directory is entirely separate from Genesis. Merge order is flexible.
• Reported lens_contents row count of 209 assumes feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 has not landed. After both PRs land, total will be 255 (Genesis 214 + Exodus 41).

Rollback plan

Revert the merge commit. The 10 new chapter files are additive and isolated under content/hermeneutic_lenses/chapters/; no existing content, schema, or app code is touched. SQLite rebuilds deterministically.

Roadmap

After this lands, batches 2–4 of Exodus complete the Exodus pilot:

• Batch 2: ex11–ex20 (Passover, Red Sea, song of Moses, manna, water from rock — major typological density expected, possibly 5+ lens average)
• Batch 3: ex21–ex30 (Book of the Covenant, tabernacle instructions — mostly legal/architectural, lower lens average)
• Batch 4: ex31–ex40 (Sabbath, golden calf, glory, tabernacle construction — golden calf chapter likely 6 lenses)

Refs #820. Closes part of #1781.

[github-actions]

Content Pipeline Results

✅ All pipeline checks passed

Step	Status	Details
Schema Validation	✅	142240 passed, 0 failed
Build DB	✅	—
DB Integrity	✅	101 passed, 0 failed

[github-actions]

Test Results

✅ All tests passed

	Passed	Failed	Total
Tests	✅ 4001	❌ 0	4001
Suites	✅ 532	❌ 0	532

Coverage

Statements	Branches	Functions	Lines
—	—	—	—

⏱️ Duration: 93.0s

[CraigBuckmaster]

Summary

Opens the Exodus pilot. Same curated-sparse cadence proven across the Genesis pilot, now applied to Exodus 1–10. 41 entries across 10 chapters, all scoring 100/100 on lens_quality_scorer after two minor restatements.

Per-chapter lens distribution

Chapter	Lenses	Notes
ex1 — Israel multiplies; midwives' courage	4	literary, redemptive, canonical, devotional
ex2 — Moses' birth and rescue	5	grammatical, literary, typological, redemptive, devotional
ex3 — Burning bush; "I AM"	6	theology peak — divine name, presence, commission
ex4 — Objections; signs; circumcision crisis	4	grammatical, redemptive, canonical, devotional
ex5 — First confrontation; bricks without straw	3	literary, redemptive, devotional — sparse, frustration
ex6 — Renewed commission; YHWH name	5	grammatical, redemptive, canonical, christocentric, devotional
ex7 — Aaron's staff; first plague	4	literary, typological, redemptive, canonical
ex8 — Plagues 2–4	3	literary, canonical, devotional — sparse, repetitive cycle
ex9 — Plagues 5–7 (incl. Rom 9:17 anchor)	4	redemptive, mission, canonical, devotional
ex10 — Plagues 8–9	3	literary, redemptive, canonical

Total: 41 entries. Average 4.1 lenses/chapter — within the curated-sparse band. ex3 maxed at 6 because it's the theological centre of the book; ex5/ex8/ex10 sparse because they are narrative repetition or frustration chapters where forcing a sixth lens would dilute.

First missional lens of the pilot: ex9. Verse 16 ("that my name may be proclaimed in all the earth") is the OT's most explicit missional charter and is cited at Rom 9:17 — earning the lens a place where Genesis only had it once (gen10/gen12).

Theological highlights

• ex3 christocentric: John 8:58's "before Abraham was, I AM" explicitly cites Ex 3:14 — and the bush that burns without being consumed (v 2) was read by Patristic readers as a type of the incarnation (presence sustaining its vessel without destroying it).
• ex2 typological: Moses saved from genocide, drawn from water, called out of Egypt — the pattern Matthew 2:13–15 explicitly applies to Christ saved from Herod and called out of Egypt. Anchored, not invented.
• ex9 mission: v 16's "that my name may be proclaimed in all the earth" → Rom 9:17 (Paul's argument from Pharaoh in his doctrine of election). Exodus's local conflict becomes the New Testament's missional theology.
• ex10 canonical: Three days of darkness (vv 22–23) anticipates Amos 8:9, Joel 2:31, and the three-hour darkness at the cross (Mt 27:45) — judgement preceding redemption is the canonical pattern.
• ex7 typological: Aaron's staff "swallows" (Hebrew בלע, bala') the magicians' (v 12) — same verb at 1 Cor 15:54 ("death swallowed up in victory"). Genuine Hebrew wordplay running across testaments.
• ex2 literary: Moses' "basket" in v 3 is Hebrew tebah — used only of Noah's ark elsewhere (Gen 6:14). The narrative literally floats salvation on the same word twice.

Pipeline gates (all green)

Gate	Result
schema_validator.py	141484 passed / 0 failed / 19 unrelated content warnings
lens_quality_scorer.py	41/41 entries at 100/100 (all four dimensions full marks)
build_sqlite.py	8 lenses, 209 contents
validate_sqlite.py	101 passed / 0 failed

Drafting discipline — lessons applied and surfaced

The 250-char ceiling held — no entry exceeded 245 chars on first draft.

Two restatements needed, both for relevance (themes-panel word overlap above the 60% threshold):

Entry	First-draft overlap	After restate
ex2 redemptive (v 24, "God remembered his covenant")	76%	<50%
ex6 grammatical (v 3, YHWH self-disclosure)	61%	<50%

Both were 93/100 on the scorer (passing the 90 floor as drafted) but lifted to 100 by switching to less-paraphrased phrasing — ex2's restatement now anchors on Hebrew zakar and parallels at Gen 8:1 / 1 Sam 1:19; ex6's now leans on yada' and the relational/informational distinction.

Pattern to flag for future Exodus batches: chapters with iconic single verses where the themes panel itself is a dense paraphrase of that verse will hit the plagiarism threshold easily. Worth building this into the drafting checklist:

• For verse anchors that are quoted in the themes panel (ex 2:24, 3:14, 6:3, 9:16, etc.), draft the entry deliberately to anchor on a different lexical handle than the themes panel uses.
• A Hebrew/Greek term, a canonical parallel, or a different verse in the same chapter all work as alternative handles.

No banned filler patterns; no should_avoid matches; no should_avoid lens-specific avoid-phrases. The lens quality cadence is healthy.

Watch list for tier-2 audit

Six entries flagged for editor scrutiny on a deeper read.

1. ex3 typological — "Patristic readers saw a type of the incarnation" (bush-not-consumed). Defensible (Gregory of Nyssa, Cyril of Alexandria, the standard Eastern reading) but brings patristic-specific imagery; some Reformed editors might prefer "anticipation" framing.
2. ex3 christocentric — Tying Jn 8:58 directly to Ex 3:14 is the standard Christological reading but compresses fast; an editor may want to note the linguistic parallel (egō eimi / ehyeh) more explicitly.
3. ex7 typological — bala' ("swallow") connecting Aaron's staff to 1 Cor 15:54 is real lexical data but is a pattern claim that some traditions would prefer as "echoes" rather than a hard typology.
4. ex9 mission — quoting v 16 then citing Rom 9:17 within a four-line entry: tightly compressed; sound but worth pacing slower in a tier-2 expansion.
5. ex10 canonical — chaining vv 22–23 → Amos 8:9 → Joel 2:31 → Mt 27:45 in one sentence is theologically defensible but covers a lot of ground; an editor may want to slow this connection.
6. ex2 typological — Mt 2:13–15 is an explicit citation; this one is on solid ground, but the framing "the pattern Matthew 2:13–15 cites" reads slightly forced. Cosmetic.

Branch / merge notes

• Branched off master at b183f7b (pre-feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796).
• PR feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 (Genesis batch 5, gen 41–50) is still in review. This PR's chapter files are non-overlapping with feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796's — Exodus directory is entirely separate from Genesis. Merge order is flexible.
• Reported lens_contents row count of 209 assumes feat(lenses): Genesis 41-50 lens content, batch 5 of pilot — closes Genesis (#820, #1781) #1796 has not landed. After both PRs land, total will be 255 (Genesis 214 + Exodus 41).

Rollback plan

Revert the merge commit. The 10 new chapter files are additive and isolated under content/hermeneutic_lenses/chapters/; no existing content, schema, or app code is touched. SQLite rebuilds deterministically.

Roadmap

After this lands, batches 2–4 of Exodus complete the Exodus pilot:

• Batch 2: ex11–ex20 (Passover, Red Sea, song of Moses, manna, water from rock — major typological density expected, possibly 5+ lens average)
• Batch 3: ex21–ex30 (Book of the Covenant, tabernacle instructions — mostly legal/architectural, lower lens average)
• Batch 4: ex31–ex40 (Sabbath, golden calf, glory, tabernacle construction — golden calf chapter likely 6 lenses)

Refs #820. Closes part of #1781.

[github-actions]

Content Pipeline Results

✅ All pipeline checks passed

Step	Status	Details
Schema Validation	✅	142240 passed, 0 failed
Build DB	✅	—
DB Integrity	✅	101 passed, 0 failed

[github-actions]

Test Results

✅ All tests passed

	Passed	Failed	Total
Tests	✅ 4001	❌ 0	4001
Suites	✅ 532	❌ 0	532

Coverage

Statements	Branches	Functions	Lines
—	—	—	—

⏱️ Duration: 93.0s