Generated from the maintained method library. Read or download equivalent Markdown.

Legant Method Library Change Log

Status / authority: Append-only change record for the client-readable methodology library
Effective date: 2026-09-03
Dependencies: README.md; LEGANT_METHOD_REFERENCE.md; REVIEW_GUIDE.md; maintainer change controls; frozen ARC release
Supersedes: Informal change notes embedded only in individual files. It does not replace source-specific development records.

List changes newest first. Historical sources and accepted decision records remain unchanged.

2026-11-02 — Public-library routing and release corrections

Approval: Chris asked to apply the reviewed corrections and publish the pending batch to Legant Labs through ChemiCloud.

Change: The AI-conversation compatibility pointer now offers a genuine request route instead of an unavailable workspace link. Internal recovery paths in this public log and pointer were replaced with maintainer-archive references; their original wording is preserved. The agent index distinguishes retired structures from safeguards explicitly retained by the Method Reference and source headers. The Method page labels its earlier Challenge/Build illustration as historical and directs new work to the nine-question Operating Manual. Generated source dates use UTC consistently. A canonical source can be explicitly excluded from a web build without deleting it or weakening the unmapped-source check.

Evidence status: Publishing, compatibility and historical-label corrections only; no new method or effectiveness claim. Academic overlap: not applicable.

Recovery: Prior sources and the original pending package are preserved in the maintainer’s dated recovery archive.

2026-11-01 — Four Narrative Constraints replaced by the nine-question Operating Manual

Approval: Chris, Nov 1: "Four Narrative Constraints --> replaced w the questions in the new 9 q&a OM." Recorded as LL-ADR-024; completes the retirement LL-ADR-022's amendment left pending.

Change: Current-authority wording now says the Operating Manual's "What I choose" answers (questions 6–9) replace the Four Narrative Constraints for new work. LEGANT_METHOD_REFERENCE.md: header, Aim sentence, §3 title and opening (the four questions kept as a legacy table; the correspondence How → Verb, Why and Personal Challenge → Contribution, Build → Sagrada, Challenge no longer required, kept), §4 inventory labels, §5 artifact row, LM-04 status, §7 source row. METHOD_ARCHITECTURE_DETAIL.md: header, refinements, Step 4 move, roles, outputs, failure mode, principle line and glossary entry, matching the Method page source. Direction/Four Narrative Constraints.md header (now legacy, with correspondence), Direction/README.md, ARC_Reference_Index.md item 3, README.md library map, Innate Edge Discovery/README.md, the Focus Line legacy header, a one-line note in Tension Framework.md, and Navigation/Client_Agent_Navigation_Standard.md change trigger. The legacy file's body is unchanged; its ownership and reasons-for-pursuit check, quality gates and failure modes still apply until a successor home is decided.

Evidence status: Status change only; no new evidence or definitions. Waypoint and Bearing are defined by their questions only.

Web copy: Stale for this change since 2026-11-01. Library and Method page rebuilt locally; not deployed.

Recovery: Prior sources are preserved in the maintainer’s dated recovery archive; request historical files from Chris.

2026-10-26 — Reusable client biographical research prompt

Approval/source: Chris’s direct request in the client research prompt conversation, 2026-10-26; accepted the concise objective-led version, added a same-name safeguard, and explicitly selected Innate Edge Discovery as the destination.

Change: Added Innate Edge Discovery/Client_Biographical_Research_Prompt.md; linked it from the root and Chris reference indexes. Reusable research input only; no methodology, skill, or historical client report changed. Academic overlap: not applicable — prompt storage and retrieval pointers.

Web copy: Publication not requested; web copy stale since 2026-10-26 for this addition. Local build only.

Recovery: Prior sources are preserved in the maintainer’s dated recovery archive; request historical files from Chris.

2026-10-26 — "Story is the current of a life" made an owned, evidence-based practice

Approval: Chris, Oct 26 (Emily Oct 31 prep thread), supplying the exact wording. Recorded as LL-ADR-023.

Change: LEGANT_METHOD_REFERENCE.md §1 principles table and METHOD_ARCHITECTURE_DETAIL.md principle statement now read: "A person's life narrative shapes their attention, interpretation, and action, so own it on purpose: choose an evidence-based story that pulls toward aligned action, and keep editing it as evidence and context change." The detail's existing paragraph (narrative connects past, present and future; consciously chosen and edited) is unchanged.

Source: Chris's post "Brainwash Yourself Into Living a Great Story" (The Zag, March 4, 2025): "Your story has got to be evidence-based, but you can cherry-pick."

Evidence status: Principle wording only; no new evidence of effect. Where the owned story lives in the nine-question Operating Manual is an open design question.

Web copy: Stale for this change since 2026-10-26. The published Method page deck line is unchanged.

Recovery: Prior sources are preserved in the maintainer’s dated recovery archive; request historical files from Chris.

2026-10-20 — Nine-question Operating Manual as the central client artifact

Approval: Chris: "The OM as centerpiece. --> I'm ready to commit." and "yeah run it with my terms for now." Recorded as LL-ADR-022.

Change: LEGANT_METHOD_REFERENCE.md §1 (Steps 3–4), §2 (Operating Manual row and a new nine-question table), §3, §5 and LM-03; METHOD_ARCHITECTURE_DETAIL.md header, framework refinements, Steps 2–4 and glossary; the published Method page's matching Step 2–4 and glossary lines. The Operating Manual now answers nine questions in two parts: About me (Core Tension, Verb, Fuel, Context, Traps), built in Step 3, and What I choose (Contribution, Sagrada, Waypoint, Bearing), recorded in Step 4. Each answer is a short top line over traceable detail; current circumstances live in a separate Baseline. A separate Four Narrative Constraints artifact becomes optional. Source of Scatter, Nozzle, Boundary Line and Archetype remain current concepts but are no longer required sections. Delivered Operating Manuals keep their structure.

Amended same day: Source of Scatter is carried by Traps; whether and where an Archetype belongs is open; active clients can be offered the nine questions. The correspondence with the Four Narrative Constraints is recorded (How → Verb; Why and Personal Challenge → Contribution; Build → Sagrada; Adversary no longer required), and the nine questions replace the Four Narrative Constraints; that retirement is pending its own update.

Evidence status: One case so far, the method owner's own; not independent. No new claim of effectiveness.

Web copy: Stale for this change since 2026-10-20. Library and Method page rebuilt locally; not deployed.

Recovery: Prior sources are preserved in the maintainer’s dated recovery archive; request historical files from Chris.

2026-10-19 — Q14 broadened to long-term favourites; duplicate Q16 resolved

Approval: Chris: "yeah update to not just be recent favorites and fix the numbering" (wording proposed in the same task).

Change: Innate Edge Discovery/Innate Edge Clues.md Q14 now asks which books, movies, shows, podcasts, YouTube channels or websites the person has loved, recently or long-term; its retelling follow-up reads "Pick your favourite story and tell it to me in your own words." The second question numbered 16 ("What are you not good at or not interested in doing professionally?") becomes a sub-bullet of Q16, and the mis-indented "Give examples" bullet is re-indented. Questions 17–31 keep their numbers, so existing references (e.g. Q31 in the decode skill) stay valid.

Evidence status: Wording and structure only; no new construct, scoring rule or validity claim.

Web copy: Stale for this change since 2026-10-19; joins the pending Q14 retelling publication. build_library.py not re-run.

Recovery: Prior sources are preserved in the maintainer’s dated recovery archive; request historical files from Chris.

2026-10-19 — Favourite-story retelling follow-up

Approval: Chris asked to add retelling to the existing book/movie question so the interviewer can notice what the participant emphasizes.

Change: Innate Edge Discovery/Innate Edge Clues.md Q14 adds: "Choose a favourite story from a book or movie and tell it to me in your own words." All other intake wording is unchanged.

Evidence status: Evidence-collection prompt only; no new construct, scoring rule, or validity claim. Academic overlap: LAM-C2, Career Construction Interview question 3; the existing research-map entry records the specific overlap.

Web copy: Stale for this change since 2026-10-19. Rebuild and publication deferred; the generated mirror is unchanged.

Recovery: The immediately preceding version is retained in Drive version history and the maintainer's dated preservation record.

2026-10-19 — Personal navigation system development note

Approval: Chris approved the note and publication to the public reference library and agent index on both hosts, excluding the Method page and main navigation.

Change: Added Personal_Navigation_System_Development_Note.md, a library entry and a Development notes category in the web library and agent index.

Authority: Emerging application and development direction; no change to canonical methodology or validated evidence claims. The approved OM example is included; personal source records remain private.

Recovery: Prior library entry, changelog and generator retained in the dated deployment record.

2026-10-08 — Governing principle added: Produce, not predict

Approval: Chris: "yes, add to the method reference." (text proposed in the same thread, built from his messages 5, 7, 9 and 15 in the Sep 29 coaching session).

Change: LEGANT_METHOD_REFERENCE.md §1 Governing principles gains a row, "Produce, not predict"; header effective-date line notes the addition. Decision record LL-ADR-021.

Evidence status: A stance on what counts as evidence of benefit, not new evidence. The two benefits are the claims already registered as unproven in §6.

Web copy: not rebuilt; joins the parked batch.

Recovery: Prior sources are preserved in the maintainer’s dated recovery archive; request historical files from Chris.

2026-09-29 — Retired DOCX intake-guide mentions

Approval: Chris: "Do you want the leftover .docx mentions removed? --> yes please".

Change: Innate Edge Clues.docx no longer exists in this library. Removed the claims that it remains a compatibility copy: Innate Edge Discovery/Innate Edge Clues.md header (Supersedes line now says the ARC-era DOCX is recoverable from the frozen release), Innate Edge Discovery/README.md (bullet removed; heading now "Convenience copy"), and LEGANT_METHOD_REFERENCE.md §4 intake paragraph (Google Doc named as the only convenience copy). Outside this library: the broken Program Docs/Innate Edge Discovery/Innate Edge Clues.docx symlink was replaced by an Innate Edge Clues.md link, and Surface Decodes/Surface_Decode_Playbook.md now points to the Markdown guide.

Evidence status: Housekeeping only; no method content changed.

Web copy: still stale since 2026-09-23 (parked batch). build_library.py not re-run.

Recovery: Prior sources are preserved in the maintainer’s dated recovery archive; request historical files from Chris.

2026-09-29 — Client self-prediction of CEO of Humanity feedback

Approval: Chris: "Innate Edge Clues.md, question 31 --> make that change" (the exact wording below, proposed in the same task).

Change: Innate Edge Discovery/Innate Edge Clues.md question 31 adds one step: before seeing anyone's answers, the client fills out the same CEO of Humanity form, answering as they think one named rater will, and enters "Self-prediction for [their name]" in the name field. No form, question, or other file changed.

Purpose: Gives a direct, field-by-field comparison between the client's self-view and an outside view, instead of relying on self-descriptions alone (self-versus-criterion discrepancy; Paulhus & Holden, 2010, Research/Other Resources/PaulhusHolden.2010.on.self-presentation.behavior.pdf).

Evidence status: Evidence-collection step only; no new method construct, scoring rule, or validity claim. With a few client-chosen raters the gap is read qualitatively, not scored.

Web copy: still stale since 2026-09-23 under the parked batched redeploy (LL-ADR-019/020); this change is added to the unshipped batch. build_library.py not re-run.

Recovery: Prior sources are preserved in the maintainer’s dated recovery archive; request historical files from Chris.

2026-09-23 — Reasons for pursuit and moral concerns as evidence

Approval: Chris: "go for it with all of r2.1. Keep it internal for now, parked for batching with other changes."

Changes: Direction/Four Narrative Constraints.md step 4 and its ownership quality gate now examine reasons for pursuing a direction, read from behaviour first and then from the person's account, with any gap treated as evidence. Innate Edge Discovery/Tension Framework.md adds "Moral concerns as evidence" (two plain-language questions per episode; MFQ-2 optional). LEGANT_METHOD_REFERENCE.md summarizes both. Outside this library: Academic Map adds C9–C12 and an inclusion rule; C6/C7 narrowed; both indexes updated.

Evidence status: Interpretive safeguards and an evidence heuristic; no new method step, validated measure, or causal claim.

Web copy: stale since 2026-09-23 by Chris's choice; parked for a batched redeploy. build_library.py not re-run.

Recovery: Prior sources are preserved in the maintainer’s dated recovery archive; request historical files from Chris.

2026-09-23 — Agent-readable web copy published

Approval: Chris approved publishing the library as plain text on legantlabs.com, repointing the Method page's library link, and noting the web copy here ("i approve both").

Purpose: Google Drive serves a viewer and sign-in page rather than file text, and blocks automated downloads, so agents following the Method page link could not read the library.

Changes: Generated a plain-text copy of the library's Markdown files at https://legantlabs.com/methodology/library/, indexed by https://legantlabs.com/llms.txt, with legacy files grouped separately. Added a web-copy note to README.md. The Method page's "Full method library" paragraph now points agents to llms.txt and keeps the Drive folder for browsing; its public wording no longer says "ARC".

Not changed: method content, statuses, or file contents in this folder beyond the README note. This folder remains the master; the web copy is regenerated from it and is not edited by hand.

Recovery: Pre-change Method page: outputs/predeploy_methodology_live_20260923.html. Build and deploy record: Legant_Labs/Legant Website/05_Method_Library_Web/ (build_library.py, MANIFEST.tsv, DEPLOY_LOG.md).

2026-09-09 — Core Tension evidence from choices and effort

Approval: Chris approved the exact proposed subsection and ledger source pointer ("yes, proceed").

Changes: Added “Choices and effort as evidence” to Innate Edge Discovery/Tension Framework.md, preserving its definition and limitations. Linked Jason’s evidence-pilot specification to the canonical subsection. Decode Phase 1.45 already requires this source; no skill or architecture change.

Evidence status: Accepted methodological heuristic; no population rarity claim or new empirical validation. Compare plausible responses, preserve alternative explanations, and test across contexts and counterexamples.

Recovery: Prior sources are preserved in the maintainer’s dated recovery archive; request historical files from Chris.

2026-09-09 — Why / Stakes beneficiaries are not limited to people

Approval: Chris's 2026-09-09 ruling during the Jason evidence-sheet definition work ("ok let's make that correction. I approve").

Purpose: The Stakes question asked what human good is at stake. Client evidence can show a person's stake in animals, places, or the natural world; the wording forced those cases into a human-beneficiary frame or excluded them.

Changes:

• Direction/Four Narrative Constraints.md: Why / Stakes question now reads "What good must be restored, protected, connected, or made possible, and for whom or for what?" Boundary note adds that the beneficiary is usually people but can be animals, places, or the natural world. Change history added.
• LEGANT_METHOD_REFERENCE.md §3: same question wording.
• Innate Edge Discovery/Tension Framework.md working terms: "valued human good" → "valued good … including who or what benefits." Change history updated.
• Root LEGANT_REFERENCE_INDEX.md Four Narrative Constraints row: "Why/Stakes = human good" → "the good at stake, for whom or what."

Not changed: the published Method page (data.json carries no "human good" wording, so no website edit is needed). No other client-readable file used the phrase.

2026-09-03 — Proportionate life-scene interpretation and narrative recompression

Approval: Chris's 2026-09-03 instruction to add the approved evidence-to-narrative boundary without creating a new ontology term or Decode workflow.

Purpose: Prevent a vivid life scene from silently proving a recurring pattern, origin, contribution, or direction while preserving the use of representative and symbolic scenes in a simple, participant-owned narrative.

Changes:

• Added one evidence-and-provenance paragraph to LEGANT_METHOD_REFERENCE.md.
• Limited closer scene analysis to load-bearing claims rather than requiring exhaustive decomposition.
• Allowed participant-facing artifacts to recompress a pattern supported across the wider record.
• Preserved the boundary between emotional or symbolic salience and evidence strength or causal proof.
• Added no framework, ontology term, required artifact, Decode phase, Four Narrative Constraints field, or operational skill change.

Recovery: Prior wording remains recoverable through file history and this dated change record. The frozen ARC release and historical client artifacts were not changed.

Validation: Confirmed the change is detailed-method wording only; existing Tension, Decode, and narrative-surface routes remain authoritative and unchanged. Public/private boundaries and current terminology were preserved.

2026-09-03 — Present-state GOAT and Stepping Stone value clarification

Approval: Chris's 2026-09-03 instruction to implement the approved waterfall architecture and explicit clarification of Stepping Stone failure, signal, and Build.

Purpose: Reconcile the Method Reference's guaranteed-accumulation wording with the published Method's ex-post compounding rule, and make the present-state GOAT comparison explicit enough to govern agent judgment.

Changes:

• Defined G.O.A.T. as the Greatest Of All Timelines still feasible from the person's actual present state, while preserving the theoretical optimum versus best-current agent hypothesis distinction.
• Clarified that every new choice begins from the new actual state; the past remains evidence and consequence, not a recoverable trajectory.
• Defined a Stepping Stone ex ante by a cost-proportionate path to useful discriminating evidence, durable contribution, or both; Step 6 determines what actually accumulated.
• Clarified that a negative external result may be useful evidence, while a costly move producing neither durable value nor useful signal is a failed stone.
• Added the boundary that later meta-learning does not retroactively justify a poorly selected move.
• Updated the published Method source, Method Reference, Four Narrative Constraints source, compiled Reference Index, and LL-ADR-017; rebuilt the generated Method page.

Recovery: Prior wording remains recoverable through file history and the dated architecture/decision records. Frozen ARC releases and historical client artifacts were not changed.

Validation: JSON/build, source consistency, root-rule parity, path, frozen-release, and independent adversarial retrieval checks are recorded in the maintainer’s dated retrieval-test record.

2026-08-29 — A centered life without continuous building

Approval: Chris's 2026-08-29 correction that a true GOAT life is not 100% dedicated to one cause, plus his instruction to fix the context and documentation so agents do not repeat the surface interpretation.

Purpose: Distinguish the organizing center of the life story from continuous Build activity.

Changes:

• Clarified that the Four Narrative Constraints give a GOAT life story its organizing center without turning every hour or activity into Build.
• Used sleep and dreaming as the simplest limit case: necessary parts of a GOAT life during which no stone is being laid.
• Preserved “one Build” as an anti-scatter constraint for contribution-bearing attempts while making clear that supporting effort, relationships, family, responsibilities, health, play, rest, curiosity, and other pursuits can sustain the person, sustain the direction, or make the life worth choosing without becoming stones.
• Added “the Build colonizes the life” as a named failure mode.
• Clarified that a non-compounding pursuit is not Source of Scatter merely because it does not add to the Build.
• Added a low-context retrieval test for the boundary and synchronized Chris's current Operating Manual and agent appendix.
• Propagated the accepted boundary into the canonical published Method content at Legant_Labs/Legant Website/02_Methodology_Page/production_build/data.json and deterministically rebuilt the generated index.html.

Recovery: Historical essays, the frozen ARC release, and prior decision records were not rewritten. The current clarification is recorded in Legant_Labs/Decision_Records.md as LL-ADR-014.

Validation: Current-source wording, cross-source terminology, Operating Manual synchronization, deterministic Method-page rebuild, and low-context retrieval canaries are checked in the dated validation packet.

2026-08-26 — Evidence-intake routing and privacy-boundary repair

Approval: Chris's 2026-08-26 instruction to make the high-priority repairs identified by the reference audit.

Purpose: Remove two ambiguous post-cutover intake routes before shared-folder readers or agents treat them as separate current methodology.

Changes:

• Migrated Innate Edge Discovery/Legant LLM Evidence Extraction Prompt.md from an ungoverned ARC-era prompt into a current optional Legant evidence-intake tool.
• Added an authority header, current Legant definitions, provisionality and evidence boundaries, and explicit participant-controlled privacy handling.
• Removed the unsupported assurance that Chris would not read the raw output, retired ARC-as-current and psych-heavy framing, and made clear that the output is not a Decode, Operating Manual, or identity verdict.
• Renamed the native Google Doc Innate Edge Interview Questions - Current to Innate Edge Interview Questions — Convenience Copy and added a visible notice that Innate Edge Clues.md remains the governing intake route.
• Updated the Discovery folder index and LEGANT_METHOD_REFERENCE.md so the governing route, optional tool, and convenience copies are distinguishable.

Recovery: Pre-repair Drive revision IDs, timestamp, size, and prompt hash are recorded in ../Methodology Releases/Legant_Library_Repair_2026-08-26/RECOVERY.md. The frozen ARC release and historical client artifacts were not changed.

Validation: PASS — authority, terminology, privacy-boundary, retrieval, Drive-title, synchronization, recovery, and frozen-release checks passed. The maintainer record is ../Methodology Releases/Legant_Library_Repair_2026-08-26/VALIDATION.md.

2026-08-05 — Constraint canon, reviewability, and legacy-source repair

Approval: Covered by the methodology owner's 2026-08-05 instruction to proceed with the audited repairs while preserving history and avoiding duplicate canon.

Purpose: Give the current Four Narrative Constraints a dedicated, challengeable source; make expert/agent review returnable; and remove confirmed current-versus-legacy contradictions without renaming or deleting historical files.

Changes:

• Added Direction/Four Narrative Constraints.md as the detailed current framework beneath the published Method page's Step 4 architecture.
• Incorporated the June 29–30 development distinctions clarified in two live direction cases: four separate layers; Challenge versus Build; Build as capability as well as artifact; one compounding build; participant ownership beyond resonance; and Stepping Stones that both confront and accumulate.
• Added framework claim IDs (FNC-01–FNC-04) with evidence boundaries and revision conditions.
• Added central method claim IDs (LM-01–LM-08) to LEGANT_METHOD_REFERENCE.md so external challenges can cite a proposition rather than the method as a whole.
• Added REVIEW_GUIDE.md with a standard human/agent review brief and return channel.
• Defined a traced maintenance rule for mapping future published essays to relevant concepts without turning publication into method authority or evidence.
• Repaired live Focus Line dependencies in the legacy-derived Archetype source without changing its historical filename or conceptual lineage; added prospective prediction and revision conditions.
• Labelled the Nozzle report sequence as historical, removed its live Focus/Boundary Line dependency, and added a prospective classification record to reduce hindsight-only validation.
• Clarified that the legacy V-V-V document routes current direction work to Four Narrative Constraints and current Core Tension boundaries to the Tension Framework.
• Kept the four audited legacy-derived or ARC-named filenames. A dependency audit found that renaming in this wave would change retrieval keys without resolving the deeper migration dependencies.
• Corrected the public legacy pointer and reader indexes so they resolve to the new governing source. Internal production routing was updated separately and remains outside this library.

Recovery: Exact pre-change files and SHA-256 hashes are preserved in a maintainer-controlled dated archive. The frozen ARC release and historical client artifacts were not changed.

Validation: See the dated maintainer validation record for authority, link, privacy, retrieval, package, snapshot, and frozen-release checks.

2026-08-04 — Legant reader-library cutover

Approval: Approved by the methodology owner on 2026-08-04.

Purpose: Make the existing canonical client-readable folder usable as the public read-only Legant Method library without creating a mirror or breaking ARC compatibility paths.

Changes:

• Replaced the stale ARC-first README.md route with the Legant Method page → Legant Method Reference → detailed-source route.
• Clarified that this folder is the single editable client-readable canon and that the website is the human discovery surface.
• Added explicit current, applied-protocol, legacy, access, exclusion, and recovery boundaries.
• Added the six-post related-writing map already curated by the published Method page to LEGANT_METHOD_REFERENCE.md.
• Added or clarified authority headers on current, legacy-derived, and legacy Markdown sources.
• Updated current navigation standards and templates from ARC/Focus Line routing to Legant/Four Narrative Constraints routing while retaining ARC for historical interpretation.
• Removed generated .DS_Store files from the share boundary.
• Moved the live backup DOCX and ungoverned Vehicle Engineering source out of the public library after verifying preservation copies in the frozen ARC release.
• Reclassified the legacy-derived positioning production source outside the public library because it contains internal production mechanics; retained the public intake artifact and documented the missing client-readable successor. Its exact destination is recorded in a maintainer-private change record.
• Converted the current Innate Edge Clues intake route to Markdown; retained the DOCX as a compatibility/download artifact.
• Anonymized client-derived examples and removed private source traces and runtime-skill details from public framework and navigation sources; full pre-public wording remains recoverable.
• Replaced the public ARC Methodology.docx with a one-page legacy access notice after the final privacy audit found identifiable client examples in the complete document. The full file remains private and checksum-verified; the internal compatibility route now resolves to the frozen copy.
• Kept the ARC Methodology.docx filename, ARC_Reference_Index.md, the folder path, and major subfolder paths in place as compatibility routes.
• Set the Google Drive folder to Anyone with the link / Viewer; Drive reports download access enabled. The parent Program Docs/ folder remains outside the public share boundary.
• Linked the public Legant Method page to this canonical reader library; the website remains the human discovery surface rather than a second methodology canon.

Recovery: Exact pre-migration sources, moved live artifacts, and their hashes remain in the maintainer's frozen ARC release and dated cleanup record outside the public share boundary.

Validation: PASS — authority, retrieval, local-link, privacy, recovery, website-deployment, and Drive-permission checks passed. The full verification record remains in the maintainer's private system.

2026-09-25 — Internal-first method ownership

Internal Method Reference now owns architecture; supporting detail preserves six steps/five alignments. Routing, review instructions and generator inputs reconciled. LL-ADR-020 records approval and recovery. Public deployment remains deferred under LL-ADR-019: web copy stale since 2026-09-23; this release adds unshipped changes on 2026-09-25. Live parity not certified.

Build note (2026-10-26, biographical prompt): local rebuild attempted but blocked by missing Python markdown dependency; no web deployment attempted. Saved prompt and both local reference pointers verified.

2026-10-28 — AI-conversation prompt release and compatibility route

Chris authorized canonical storage and reference mapping of the two-prompt participant-trial package. The former V24 entry is a compatibility pointer to the current package, available from Chris. Personal teaching examples remain outside the public-library build. No method architecture or prompt behavior changed in this release; Academic overlap: not applicable (routing/preservation only). Prior V24 and indexes are preserved in the maintainer’s dated recovery archive. Publication was deferred at the time; the local build/check result was recorded in the maintainer’s release verification and website deployment log.