Adopt SYN as the eighth ORA entry type

Adopt SYN as the eighth ORA entry type

The decision

Codex adopted SYN as a first-class ORA entry type on 2026-04-28, updating the content model from v1.0 to v1.1.

Context

The live ORA tree already contained docs/entries/syntheses/, and the entry ID guard already treated that directory as a canonical ORA entry surface. The written content model and README still described seven types and omitted syntheses from the layout and quick-start instructions. That created two maps for authors: the tooling said syntheses were entries, while the documentation said they did not exist.

Three existing syntheses showed that the type was not incidental: 1. ORA-2026-0076 integrates fleet-antagonism dynamics and maps related doctrine candidates. 2. ORA-2026-0104 integrates epistemic flatness, dissolution, and world-model fidelity. 3. ORA-2026-0105 integrates Gemini sentinel-loop architecture and persistent self-direction.

Alternatives considered

Option A: Make syntheses first-class entries — CHOSEN

• Pros: Aligns docs with tooling and existing canonical files; gives cross-entry integration a clear home; avoids forcing synthesis prose into patterns, decisions, or doctrines before it has that shape.
• Cons: Adds one more type for authors to learn.
• Why we picked it: The type already exists in practice. Naming it makes the map honest and gives authors a lower-friction place to integrate multiple sources without prematurely promoting evidence.

Option B: Remove docs/entries/syntheses/ from the guard

• Pros: Keeps the original seven-type model intact.
• Cons: Demotes existing synthesis entries into an ambiguous artifact surface even though they carry ORA IDs, frontmatter, sources, status, and maturity.
• Why we rejected it: The files are already functioning as ORA entries. Removing them from guard coverage would make identity drift easier, not harder.

Option C: Force each synthesis into an existing type

• Pros: Avoids a new code.
• Cons: Misclassifies integrative frames as patterns, decisions, or doctrines. That muddies the evidence ladder and encourages premature promotion.
• Why we rejected it: Synthesis is connective tissue. It may later support a PAT, HYP, EXP, DOC, or DEC, but it should not pretend to be one.

Decided by

• Decision maker: CODEX-CLI-Chad-Barlow-iMac-ORA-z8ir while executing ORA-1899.
• Input from: existing guarded syntheses entries and the ORA content-model invariant that model changes must be tracked as decisions.
• Opposed by: none recorded.

Implications

• Creates: SYN as an allowed ORA entry type, docs/schema/templates/synthesis.md, and explicit README layout for entries/syntheses/.
• Kills: The implicit idea that syntheses are stray prose outside the structured content model.
• Affects: Future ORA authors can use SYN when integrating multiple entries or artifacts into one frame, while leaving evidence promotion to PAT/HYP/EXP/DOC/DEC entries.

Sunset / revisit

• Revisit trigger: If syntheses become a dumping ground for uncategorized claims rather than integrations across named sources, split or tighten the type.
• Hard sunset: none.