Research journal — FE-CFDP-CLAUDE-001

Objective

First real (non-synthetic) execution of the ROS–FE profile, in the framework-discovery-agent role, targeting the OODA Loop (Observe–Orient–Decide–Act). This is the "one real, source-preserving provider execution" that FE-CFDP's next_execution field called for.

Scope and assumptions

• Provider: claude (this session), declared explicitly by the user rather than inferred.
• Research question and role were not supplied with the initial instruction to run start-work-agent.md; both were confirmed with the user via a clarifying question before work began, since guessing them would have materially changed the research scope.
• Target framework (OODA Loop) was one of two options offered from the active FE-008 Reference Corpus Batch 001 list in the main repo's CURRENT_STATE.md; the user selected it.

Write-boundary decision

START-HERE.md and execution-profile.md describe canonical records living under ros-profile/records/, but the synthetic execution's own records/README.md states: "A real execution keeps draft records here until reviewed and registered." Because the user's authorization covered running this execution, not profile maintenance, all records produced here are held as drafts inside this execution directory (records/sources/, records/frameworks/, records/genomes/, records/observations/, records/contradictions/) rather than written into the canonical ros-profile/records/ tree. This keeps them outside ros-fe:validate/ros-fe:registry scope until a separate, explicitly authorized promotion step reviews and registers them.

Source acquisition

Used WebSearch/WebFetch to identify and bibliographically check five sources:

• FE-SRC-OODA-001 — Boyd, "Patterns of Conflict" (Dec 1986 briefing edition) — primary.
• FE-SRC-OODA-002 — Boyd, "The Essence of Winning and Losing" (1995–96 briefing) — primary; reportedly the only diagram of the loop Boyd himself drew.
• FE-SRC-OODA-003 — Osinga, Science, Strategy and War: The Strategic Theory of John Boyd (2007) — secondary scholarly.
• FE-SRC-OODA-004 — Hankins, "The Korean War and the OODA Loop: What Happened to the Kill Ratio?" (2025) — secondary critical commentary.
• FE-SRC-OODA-005 — Wikipedia "OODA loop" article — tertiary, used only for orientation.

Neither primary briefing (FE-SRC-OODA-001, FE-SRC-OODA-002) was fetched and read in full; claims attributed to them are limited to what secondary-source summaries or search-result excerpts corroborate. This is recorded as a limitation on every downstream record rather than treated as full-text verification.

Findings worth flagging

• The four-box cyclical diagram in near-universal popular circulation is not Boyd's own diagram; his one self-drawn sketch is reported to be more elaborate, with a heavily developed Orient stage and feed-forward/feedback paths. Recorded as FE-OBS-OODA-001.
• A commonly cited ~10:1 Korean War kill-ratio figure used to support the framework's tempo thesis is disputed by a 2025 critical source, and this execution could not independently confirm whether Boyd himself originated that specific figure. Recorded as a genuine, unresolved contradiction — FE-CONTRA-OODA-001 — rather than resolved in either direction.
• No formal decision rule for the "Decide" stage, and no adaptation/tailoring guidance, were found in the primary sources reviewed; both are marked not-reported rather than inferred.

What was not done

• No claim about the OODA loop's effectiveness, superiority, or empirical validation was made or implied.
• No canonical FE theory was touched.
• No second-provider replication was attempted (single-provider execution; role separation would not constitute replication per FE-CFDP's prohibited_claims).