eval-results template: Transferable Lessons + Proposed Changes (Loop Inputs) sections

- _templates/eval-results.md: new Transferable Lessons section (2-5 generalizable
  bullets + mandatory methodology-note promotion check) and Proposed Changes (Loop
  Inputs) section (concrete candidate wording, tagged by target surface and measured
  gap — the tuning-loop handoff artifact).
- os-vault-write-eval-baseline-grid-results.md: backfilled to the new template.
  Four transferable lessons (headline: persistence evals need a destination axis,
  not just a trigger axis) and four verbatim WS2 loop-input candidates: destination
  ladder wording, query-side triggers (no eval coverage yet), query-first
  update-vs-create rule, type-misfit routing. No os-vault skill/hook surface touched
  — applying candidates starts the wording loop.
- eval-methodology-ladder.md: documented the lesson-promotion pattern from the
  methodology side.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

_templates/eval-results.md

@@ -48,11 +48,22 @@ source: [project name]  # project that contains the eval harness
 **Weaknesses of This Eval (Its Ladder Level)**
 <!-- What this eval can't see or didn't test. Examples: single fixture generalization (would a second fixture in a different language / domain change the results?), 1 rep/cell variance (high noise floor), no ambiguity axis (scenarios are clear-cue vs no clear boundary-case testing), limited distractor count, no longer held-out (wording was tuned against this grid — it's now training-set), model-specific failure modes, ablation surfaces that were never tested (e.g., "channel ablation not run — don't assume hook redundancy"). Open questions the eval can't resolve. Open questions the eval raises. -->
 
+## Transferable Lessons
+
+<!-- 2–5 bullets MAX. Each bullet is a rule someone designing a DIFFERENT eval could apply, phrased as a generalizable statement with a one-line why. NOT a recap of this eval's results — only things that transfer (process mistakes to avoid, harness-design gotchas, phrasing mechanisms that worked). If a lesson is recurring across evals, it belongs in a methodology note, not here — see the promotion prompt below. -->
+
+- [Generalizable rule — why it matters]
+
+**Promotion check:** <!-- Required before finishing this note: does any lesson above belong in a methodology note ([[eval-methodology-ladder]], [[running-autoresearch-skill-evals]], or a new one)? If yes, update that note NOW and wikilink it here; the lesson stays here as a bullet pointing at the methodology note. Eval-results notes are episodic records — methodology notes are where compounding knowledge accretes. State the outcome: "promoted X to [[note]]" or "nothing recurring yet." -->
+
 ## Deployment and Evolution
 
 **Good-Enough Gate**
 <!-- Explicit criterion under which these results justify deployment, adoption, or real-world rollout. Examples: "Sonnet 8/8 passes for pilot rollout on projects with Rust / Go codebases; haiku pending 3-rep confirmation on the W3 edge case (running today, 2026-07-06)." Or: "5/8 passes the 'prompting issue, not capability gap' threshold; spin up a wording loop before production rollout." Tier-specific status. Conditions that would change the call. -->
 
+**Proposed Changes (Loop Inputs)**
+<!-- If these results motivate changing anything (skill wording, hook text, CLAUDE.md sections, harness criteria), write the CONCRETE candidate text here — the actual proposed wording in clear language, not a description of it. This is the handoff artifact to the tuning loop or follow-up session: a reader (human or AI) should see exactly what is proposed to change and why, and later diff it against what actually shipped. Tag each candidate with its target surface (which file/section) and which measured gap it addresses. If nothing is proposed, say "none." -->
+
 **Hardening Path / Next Measurement**
 <!-- If there is a follow-up eval harness (e.g., Eval C at a higher ladder level), point to it. Otherwise, point to the ladder-approach methodology note explaining the progression. One paragraph or brief list of what gets tested next and why (e.g., "Eval C will add ambiguity-ladder discrimination scenarios to test whether the model mistakes clear-cue for ambiguous-cue; if it does, wording tuning stops and the feature is capability-limited"). -->
 

eval-methodology-ladder.md

@@ -9,7 +9,7 @@ tags:
   - convention/eval-design
   - tool/autoresearch
 scope: global
-last_updated: 2026-07-06  # hardened same day with Eval C first-grid lessons (checker-conformance, self-test blind spot)
+last_updated: 2026-07-07  # added lesson-promotion pattern (eval-results → methodology notes, Loop Inputs handoff)
 date: 2026-07-06
 related:
   - running-autoresearch-skill-evals
@@ -72,6 +72,17 @@ Run each scenario multiple times (reps) at a given level. Reps catch variance an
 
 Lower model tiers (e.g., haiku vs sonnet) may not clear the same rung at the same pass bar. E.g., "sonnet clears level 1 at 3-rep majority (all scenarios PASS), so move to level 2; haiku still at 2/3 on some scenarios, so continue level 1 with more reps." The ladder is flexible per-tier.
 
+### Pattern: Lesson promotion from eval-results notes
+
+Eval-results notes are episodic records of one measurement; methodology notes (this one and
+its siblings) are where compounding knowledge accretes. The `eval-results` template ends with
+a mandatory promotion check: any transferable lesson that recurs (or names a new confound)
+gets written into the relevant methodology note at the moment the eval-results note is
+finished — not left stranded per-eval. The eval-results note keeps a one-line bullet pointing
+at the methodology note. Eval-results notes also carry a **Proposed Changes (Loop Inputs)**
+section holding the *concrete candidate wording* a tuning loop will test — the handoff
+artifact that lets a later reader diff proposal against what shipped.
+
 ### Pattern: Stop criterion
 
 **"Good enough for now → test IRL"** is the gate between ladder evals and production validation. When the current level is reliably PASS and you're confident in the behavior, stop optimizing for that level and prepare for production rollout. That gate is only meaningful if **deliberate, instrumented observation** of real sessions (see [[eval-methodology-irl-feedback-loop]]) will surface silent misses and feed them back as new scenarios.

os-vault-write-eval-baseline-grid-results.md

@@ -9,7 +9,7 @@ tags:
   - tool/claude-code
   - project/cc-os
 scope: global
-last_updated: 2026-07-06
+last_updated: 2026-07-07  # added Transferable Lessons + Proposed Changes (WS2 Loop Inputs) sections
 date: 2026-07-06
 related:
   - eval-methodology-ladder
@@ -85,12 +85,71 @@ only via the informational reasons trail, not a dedicated axis; no measurement o
 `/os-vault:write` skill when explicitly invoked (that's an Eval-A-shaped question, out of
 scope here).
 
+## Transferable Lessons
+
+- **Persistence evals need a destination axis, not just a trigger axis.** When the harness (or any host tool) has a competing built-in persistence channel, "did the model save it?" and "did it save it to the right place?" are separate failure modes — this grid would have read as a flat trigger gap without the reasons-trail showing every L1 rep wrote to auto-memory.
+- **"Remember this" wording loses to the default channel.** Trigger wording that names the action but not the destination drains to whatever persistence path is cheapest/built-in. Candidate wording must name the destination explicitly.
+- **Count canary reps.** The one vault-conforming pass was a canary; had it been excluded, the ceiling-exists evidence would have been lost (and including it avoids peeking bias in the denominator).
+- **A clean-negative baseline converts to a hard regression gate.** 18/18 negatives means the tuning loop gets a free, unambiguous stop condition: any negative FAIL during wording iteration is a hard stop, no judgment needed.
+
+**Promotion check:** the destination-axis lesson (first bullet) is a new eval-design confound not yet covered by [[eval-methodology-ladder]] — promote it there if a second eval hits a competing-channel shape; nothing else recurring yet. "Count canary reps" is already practiced in [[running-autoresearch-skill-evals]] discipline.
+
 ## Deployment and Evolution
 
 **Good-Enough Gate** — Not applicable yet; this grid establishes the gap the WS2 wording
 loop must close. Negatives at 18/18 mean restraint needs no work — wording iteration should
 optimize trigger + routing only, and any future negative regression is a hard stop.
 
+**Proposed Changes (WS2 Loop Inputs)** — concrete candidate wording drafted 2026-07-07,
+NOT yet applied to any os-vault surface (applying it starts the wording loop and converts the
+run-set to training-set; measurement then moves to the frozen reserve). Diff these against
+what actually ships.
+
+*Candidate 1 — destination ladder* (targets: write SKILL.md description/body + an os-vault
+SessionStart usage note + candidate trigger-phrased CLAUDE.md section; addresses both measured
+gaps — L1 mis-routing and L2/L3 under-trigger):
+
+> **Where knowledge goes — ask in order:**
+> 1. Does the repo already record it (code, ADR, CLAUDE.md, git history)? → write it there or
+>    nowhere; never duplicate into memory.
+> 2. Would this change how you act in a **different** repo, next year? → the **SecondBrain
+>    vault** via `/os-vault:write` — NOT auto-memory. Facts about the world: tool behavior,
+>    client conventions, API quirks, methodology that worked. The repo is where you *learned*
+>    it, not what it's *about*.
+> 3. Only useful in future sessions of **this** repo (where things live here, what we tried
+>    here)? → auto-memory.
+> 4. Task status, session state, in-flight progress → nothing.
+>
+> Tiebreaker: what is the fact **about**? About a tool/client/API → vault. About this repo →
+> auto-memory. When a durable fact surfaces mid-task — even with no one saying "remember" —
+> run the ladder before moving on.
+
+*Candidate 2 — query-side triggers* (targets: query SKILL.md description + SessionStart usage
+note; NO eval coverage yet — a query-side eval is a separate future harness, don't score it
+against this one):
+
+> - Before designing or setting up an eval → `/os-vault:query` for `type/eval-results` +
+>   `domain/llm-evaluation` (and the project facet): what was already run, what the
+>   hypotheses/results were, what mistakes not to repeat.
+> - Before first use this session of a tool you have vault notes for → query `tool/<name>`.
+> - Before starting work involving a client → query `client/<name>`.
+> - Don't re-query a facet already queried this session — the note content is in context.
+
+*Candidate 3 — update-vs-create rule* (target: write SKILL.md procedure):
+
+> Write is query-first: before creating a note, query for existing notes on the same subject.
+> New fact contradicts/extends/refines an existing note → update it and bump `last_updated`.
+> Genuinely new subject → new note.
+
+*Candidate 4 — type-misfit routing* (target: write SKILL.md; conservative, human-gated
+taxonomy growth):
+
+> Content that fits no existing note type: do NOT invent a type or force a bad fit silently.
+> Write it under the closest type with a visible `## Type-fit note` line saying what didn't
+> fit, or — if the shape is clearly recurring — propose a new type via
+> `/os-vault:design-template`. Misfits get reviewed in `/os-vault:reorganize`; three misfits
+> of the same shape justify a new type.
+
 **Hardening Path / Next Measurement** — Per `ws2-os-vault-write-eval.md` step 4: an
 `/autoresearch` wording loop (checker/fixtures/scenarios/rubric frozen; surfaces = write
 SKILL.md description/body, an os-vault SessionStart usage note, possibly a trigger-phrased