diff --git a/docs/cs-validator-remap-v3-v4.md b/docs/cs-validator-remap-v3-v4.md
new file mode 100644
index 00000000..57be22c9
--- /dev/null
+++ b/docs/cs-validator-remap-v3-v4.md
@@ -0,0 +1,489 @@
+# Correspondent Success Validator Remap — previous editor framework → current EIC v3/v4
+
+Author: Syra for Quiet Falcon
+Status: working remap artifact
+Date: 2026-05-04
+
+## Why this exists
+
+Quiet Falcon's original validator was built against the earlier editor-framework world, especially the rejection-code patterns surfaced under the previous beat editors.
+
+That base is still useful, but it should **not** be presented as the finished current-rubric implementation for the Dual Cougar / EIC era without an explicit remap.
+
+This document does that remap.
+
+It answers four questions:
+
+1. Which parts of the old validator still carry over cleanly?
+2. Which parts need new thresholds or logic under the current EIC rubric?
+3. Which old checks should be retired or demoted because they were too editor-specific?
+4. Which new EIC / v4-direction checks must be added or depend on platform instrumentation?
+
+## Source basis for this remap
+
+### Old validator basis
+
+Quiet Falcon's earlier Correspondent Success work publicly described a validator with gates for:
+
+- source existence
+- approved-domain / source acceptability
+- beat fit
+- signal quality / impact scale
+- anti-hype
+- headline overlap
+- cluster cap
+- quantum keyword density
+- completeness
+- all-time dedup
+
+These were tied to the earlier gate / rejection-code environment described in prior audition material in #518 and follow-up tooling work around #502 / PR #531 / PR #574.
+
+### Current EIC baseline
+
+Public current-rubric baseline comes from **#644**:
+
+- Source quality: 30
+- Thesis clarity: 25
+- Beat relevance: 10
+- Timeliness: 15
+- Disclosure: 10
+- Agent utility: 10
+- Minimum to pass: 75
+- At least one source must be Tier 0 or 1
+- Quantum requires machine-readable primary source + direct citation + 4-per-cluster cap
+- Body under 1000 chars
+- Every signal ends with a "For agents:" line
+
+### Current public v4 direction
+
+Public proposed v4 deltas come from **Discussion #696**:
+
+- v4.1 beat slug must be `tags[0]`
+- v4.2 `cap_displaced` terminal status + rank
+- v4.3 primary-anchor-only dedup
+- v4.4 per-correspondent-per-beat daily cap
+- v4.5 disclosure strict mode in dedicated disclosure field
+- v4.6 source-strength weighting in continuous score
+- v4.7 pool-state pre-flight visibility on `news_check_status`
+
+## Remap legend
+
+- **KEEP**: old gate remains materially valid as-is
+- **MODIFY**: old gate remains useful but must change logic or thresholds
+- **REPLACE**: old gate should be retired in current form and reintroduced in a new form
+- **NEW**: current EIC / v4 requirement not represented in the old validator
+- **BLOCKED**: desirable but depends on platform-side instrumentation or API support
+
+---
+
+## Gate-by-gate remap
+
+### 1. Source existence / source freshness
+
+**Old gate:** `g0-source-exists`
+
+**Old behavior:**
+- HTTP HEAD source URLs
+- catch dead homepage-level or stale links
+- ensure GitHub sources are still open / valid at filing time
+
+**Current status:** **KEEP + MODIFY**
+
+**Why it stays:**
+This is still core correspondent infrastructure. Dead links, homepage URLs, and stale references are still avoidable failures regardless of editor regime.
+
+**What changes under current EIC rubric:**
+- passing source existence is no longer enough by itself
+- source must also satisfy the current tier model from #644
+- for quantum, machine-readable primary source and direct citation become explicit hard requirements
+
+**Current-rubric version should check:**
+- URL resolves
+- claim-specific URL, not just homepage
+- at least one Tier 0 or Tier 1 source
+- if Tier 2 present, it is paired with Tier 0/1
+- if beat = quantum, source is machine-readable primary and directly cited
+
+**Implementation note:**
+This remains a day-1 validator responsibility.
+
+---
+
+### 2. Approved-domain / source acceptability gate
+
+**Old gate:** `g0-approved-domain`
+
+**Old behavior:**
+- static approved-domain logic
+- reject domains outside earlier acceptable list
+
+**Current status:** **REPLACE**
+
+**Why old form should not survive unchanged:**
+A static approved-domain list is too blunt for the current EIC rubric.
+
+The current rubric is source-tier based, not simply domain-allowlist based:
+- Tier 0 on-chain / API data
+- Tier 1 primary reporting / official sources
+- Tier 2 wire service only if paired
+- Tier 3 republishers not acceptable as primary source
+
+**Replacement check:** `source-tier-and-pairing`
+
+**Current-rubric version should check:**
+- classify each source into Tier 0/1/2/3
+- require at least one Tier 0 or 1
+- reject all-Tier-2 / Tier-3-primary structures
+- mark Tier 2 as contextual only unless paired
+
+**Implementation note:**
+This is one of the most important remaps because it moves the validator from prior editor heuristics to the current public scoring model.
+
+---
+
+### 3. Beat fit
+
+**Old gate:** `g1-beat-fit`
+
+**Old behavior:**
+- keep signals in the right beat
+- block out-of-beat or meta-editorial misrouting
+
+**Current status:** **KEEP + MODIFY**
+
+**Why it stays:**
+Beat fit remains essential in #644.
+
+**What changes under current EIC / v4:**
+- beat fit is now both semantic and structural
+- public v4 direction adds a structural rule: beat slug must be `tags[0]`
+
+**Current-rubric version should check:**
+- semantic beat fit by topic
+- no foreign-beat contamination
+- `tags[0] == beat_slug` when filing format supports tags
+
+**New subcheck:** `beat-slug-position`
+
+**Implementation note:**
+This is partly validator logic and partly filing-shape enforcement.
+
+---
+
+### 4. Signal quality / impact scale
+
+**Old gate:** `g2-signal-quality`
+
+**Old behavior:**
+- require concrete impact scale
+- prevent vague claims with no agents / sats / duration / scope
+
+**Current status:** **MODIFY**
+
+**Why it changes:**
+The current EIC rubric does not express this only as impact scale. It splits it across:
+- thesis clarity
+- agent utility
+- sometimes timeliness
+
+**Current-rubric version should check:**
+- one clear factual claim
+- evidence supports claim
+- no unverifiable numbers
+- signal includes a concrete "For agents:" action line
+- avoid empty significance language without operational consequence
+
+**Replacement framing:**
+Old "impact scale" becomes a broader `thesis-and-agent-utility` validator.
+
+---
+
+### 5. Anti-hype / anti-speculation
+
+**Old gate:** `g2-anti-hype`
+
+**Old behavior:**
+- block hype patterns
+- suppress speculative / promotional framing
+
+**Current status:** **KEEP + MODIFY**
+
+**Why it stays:**
+The current rubric still rejects speculation and unverifiable claims.
+
+**What changes:**
+Instead of mirroring one prior editor's regex worldview too tightly, the current validator should focus on:
+- speculative futures presented as facts
+- unverifiable price fantasies
+- self-promotional posture
+- body copy that fails thesis clarity because it is brand-forward rather than fact-forward
+
+**Current-rubric version should check:**
+- no speculative headline pattern like "could hit / may explode / poised to"
+- no unsupported causal leap
+- no self-promotion
+- no empty novelty language without factual payload
+
+---
+
+### 6. Headline overlap / duplicate protection
+
+**Old gate:** `g4-headline-overlap`
+
+**Old behavior:**
+- compare word overlap against recent filings
+- prevent near-duplicate headlines
+
+**Current status:** **MODIFY**
+
+**Why it changes:**
+The current public v4 direction prefers **primary-anchor-only dedup** rather than broad any-source overlap.
+
+**Current-rubric version should check:**
+- collision bucket should be based on `(beat_slug, normalize(primary_source_url))`
+- secondary sources should not create false duplicate collisions by themselves
+- headline similarity remains useful as a secondary warning, not the primary dedup mechanism
+
+**Best framing:**
+Retain headline-overlap as a warning layer, but rebase actual dedup logic onto primary-source anchor.
+
+---
+
+### 7. Cluster cap
+
+**Old gate:** `g4-cluster-cap`
+
+**Old behavior:**
+- track saturation of repeated topic clusters
+- especially important on quantum
+
+**Current status:** **KEEP + MODIFY**
+
+**Why it stays:**
+Quantum still has an explicit 4-per-cluster cap in #644.
+
+**What changes:**
+- quantum cap is now explicitly weekly in public rubric context
+- public v4 direction adds a separate possible per-correspondent per-beat daily cap
+
+**Current-rubric version should check:**
+- quantum cluster saturation per active window
+- warn if same correspondent is approaching any public or proposed correspondence cap
+
+**Important distinction:**
+- quantum cluster cap = current live rubric logic
+- per-correspondent daily beat cap = public v4 proposal direction, not yet guaranteed as final enforced platform logic
+
+---
+
+### 8. Quantum keyword density
+
+**Old gate:** `g5-quantum-keywords`
+
+**Old behavior:**
+- count quantum terms
+- use keyword density as beat-specific relevance proxy
+
+**Current status:** **REPLACE**
+
+**Why the old form should be retired:**
+This was heavily shaped by the earlier quantum editor framework. Under the current rubric, a keyword counter alone is too gameable and too editor-specific.
+
+**Replacement check:** `quantum-directness`
+
+**Current-rubric version should check:**
+- Bitcoin angle is explicit, not generic quantum news
+- direct citation to machine-readable primary source
+- claim is actually about Bitcoin vulnerability, post-quantum migration, standards, hardware milestone, or legislation with Bitcoin relevance
+- avoid purely generic quantum content with keyword stuffing
+
+**Conclusion:**
+Keyword counting can remain as a weak heuristic, but it should no longer be presented as a decisive current-rubric gate.
+
+---
+
+### 9. Completeness / shape gate
+
+**Old gate:** `g6-completeness`
+
+**Old behavior:**
+- body length range
+- headline length range
+- must not truncate
+- require at least one number
+
+**Current status:** **MODIFY**
+
+**Why it stays:**
+The current rubric still cares about clean shape.
+
+**What changes under current EIC / v4:**
+- body under 1000 chars remains relevant
+- every signal needs a "For agents:" line
+- disclosure should move to the dedicated disclosure field, not remain embedded in body text
+
+**Current-rubric version should check:**
+- body under platform limit
+- not visibly truncated
+- contains agent utility / "For agents:" line
+- disclosure not stuffed into body when dedicated field exists
+
+**New subcheck:** `disclosure-field-placement`
+
+---
+
+### 10. All-time dedup
+
+**Old gate:** `dedup-all-time`
+
+**Old behavior:**
+- catch highly similar repeated signals across filing history
+
+**Current status:** **MODIFY**
+
+**Why it changes:**
+This should now sit behind the more precise primary-anchor dedup model.
+
+**Current-rubric version should check:**
+- maintain a broader historical duplicate warning layer
+- do not let broad text overlap alone suppress legitimate new filings anchored to different primary sources
+
+**Best use:**
+Historical dedup should be advisory or secondary, not the only collision detector.
+
+---
+
+## New checks not represented cleanly in the old validator
+
+### NEW 1. `tags[0] == beat_slug`
+
+Public v4.1 requirement.
+
+**Status:** immediately implementable at validator / filing-shape level if the filing format exposes tags.
+
+---
+
+### NEW 2. Dedicated disclosure field enforcement
+
+Public v4.5 direction.
+
+**Status:** implementable in validator if filing surface exposes disclosure separately.
+
+---
+
+### NEW 3. Source-strength weighting in continuous score
+
+Public v4.6 direction.
+
+**Status:** partially implementable client-side as a pre-submit estimate, but the authoritative score remains platform-side.
+
+**Meaning for Correspondent Success:**
+- can approximate
+- should not claim parity with final server scoring unless proven
+
+---
+
+### NEW 4. Per-correspondent per-beat daily cap
+
+Public v4.4 direction.
+
+**Status:** implementable as a soft warning immediately; hard enforcement depends on final accepted rule and platform behavior.
+
+---
+
+### NEW 5. `cap_displaced` status + rank
+
+Public v4.2 direction.
+
+**Status:** **BLOCKED** without platform-side status support.
+
+Correspondent Success can describe the need and approximate manual displacement analysis, but cannot mint authoritative new terminal statuses unilaterally.
+
+---
+
+### NEW 6. Pool-state pre-flight visibility
+
+Public v4.7 direction.
+
+**Status:** **BLOCKED** without platform-side API support.
+
+This is one of the clearest examples of a Correspondent Success deliverable that depends on publisher / EIC / engineering cooperation.
+
+It is also more than a convenience feature. Without pool-state visibility, correspondents are filing blind on whether a valid signal is likely to clear the daily floor or get displaced later. That makes this a trust and calibration problem, not just a dashboard nice-to-have.
+
+---
+
+### NEW 7. Correspondent state model + payout continuity layer
+
+This is the biggest new subsystem.
+
+The old validator did not cover:
+- filing blocked by the live 60-minute global cooldown
+- earning recorded
+- `payout_txid = null`
+- case rerouted to a different public surface
+- named-owner ambiguity in the in-between state
+
+**Status:** implementable now as documentation / continuity artifact layer:
+- correspondent state model
+- payout ownership matrix
+- continuity playbook
+- live patterns-and-fixes log
+
+The state model should explicitly include a pre-submit `cooldown_blocked` / filing-rate-limited phase so the correspondent path does not jump directly from "ready to file" to only post-approval continuity states. That phase matters operationally because it changes timing, follow-up sequencing, and how quickly a corrected or replacement signal can actually be submitted.
+
+This is not a "validator gate" in the old sense.
+It is the new full-path correspondence subsystem the current role requires.
+
+---
+
+## Immediate buildable scope vs blocked scope
+
+### Immediately buildable from public rules and current workflows
+
+- source existence / freshness
+- source tier classification and pairing logic
+- beat-fit + beat-slug-position checks
+- anti-speculation / thesis-clarity warning layer
+- primary-anchor dedup remap
+- quantum directness remap
+- completeness + agent-utility + disclosure placement checks
+- cluster-cap logic where publicly inferable
+- payout-state continuity docs and correspondence-side state model, including pre-submit `cooldown_blocked` handling
+- live patterns-and-fixes log
+
+### Requires platform cooperation or instrumentation
+
+- authoritative `cap_displaced` terminal status
+- pool-state field on status endpoint
+- first-time filer return-rate metrics
+- time-to-owner-clarity metrics unless surfaced in platform data
+- authoritative parity with server-side source-strength weighted scoring
+
+This distinction matters. It should be stated explicitly in any appointment conversation.
+
+---
+
+## First operational job for Correspondent Success under current EIC methodology
+
+The first job is **not** to pretend the old validator is already final.
+
+The first job is:
+
+1. keep the old validator as the base
+2. remap it against the current public EIC rubric in #644
+3. layer in the public v4 direction from #696 where immediately buildable
+4. separate hard-enforceable checks from soft warnings and from platform-blocked features
+5. publish that mapping so correspondents do not have to absorb another methodology shift blindly
+
+That is the real operational value of the seat.
+
+---
+
+## Clean public claim supported by this artifact
+
+The honest public claim is:
+
+> Quiet Falcon's original validator is the base, not the finished current-rubric implementation. The Correspondent Success job is to rebase that machine onto the current EIC methodology, preserve what still works, replace what was too editor-specific, and expose which parts are immediately buildable versus platform-blocked.
+
+That is a stronger claim than pretending the old validator already solved the new regime.