diff --git a/.github b/.github
new file mode 100644
--- /dev/null
+++ b/.github
@@ -0,0 +1,1 @@
+/home/me/aesthetic-computer/modes
\ No newline at end of file
diff --git a/grants/SCORE.md b/grants/SCORE.md
--- a/grants/SCORE.md
+++ b/grants/SCORE.md
@@ -7,12 +7,18 @@
| Application | Deadline | Amount | Status | Folder |
|---|---|---|---|---|
| Error 406 [Netstalgia] Not Acceptable | 2026-05-04 (23:59 CEST) | €1,500–7,000 (asking €5.5k) | DRAFT v0 ready | `error-406-netstalgia/` |
-| NY Art Book Fair — Project Space | 2026-05-08 | self-funded (Project Space fee covered by Aesthetic Inc) | GO — public preview live at [aesthetic.computer/printed-matter-2026](https://aesthetic.computer/printed-matter-2026) | `printed-matter-project-space-2026/` |
+| NY Art Book Fair — Project Space | 2026-05-08 | $3,500 stipend | DRAFT v0 ready | `printed-matter-project-space-2026/` |
| Oracle Egg — BROILER Residency (Fall 2026) | 2026-05-15 (23:59 PST) | TBD | DRAFT v0 ready | `oracle-egg-broiler-2026/` |
+<<<<<<< HEAD
| Serpentine — Future Art Ecosystems R&D Fellowship: Art × Convergence | 2026-06-07 (midnight BST) | £10,000 + London travel/lodging | DRAFT v0 ready | `serpentine-fae-2026/` |
| SSRC — Just Tech Fellowship | 2026-06-28 (23:59 EST) | up to $60,000 (1 yr, unrestricted) | SOURCED — researching (Fia, 2026-05-14); no folder yet | — |
| ~~CalArts CCAT — Technical Director, BB6 studio (job, not grant)~~ | posting closed (verified 2026-05-21) | $100k–$125k / yr | **CLOSED — not submitted.** Paycom job 154720 returns "We Couldn't Find This Job"; CalArts careers portal lists no CCAT roles. Verified: CCAT = CHANEL Center for Artists and Technology, ED MaryClare Brzytwa. Draft archived in vault `calarts-ccat-tech-director-2026/`; public page reframed as archived record. Reopen if CCAT reposts a TD / related role. | `calarts-ccat-2026/` |
| ~~Francis Gallery painting exhibition~~ | passed (2026-04-30) | two $15k prizes | MISSED — the public Summer Exhibition deadline was 2026-04-30 (Fia's text said May 30, incorrect). Track 2027; if a separate private call exists, restart this folder. | `francis-gallery-2026/` |
+=======
+| CultureHub — LA Residency 2026 | 2026-05-20 (23:59 PT) | $2,000 stipend | DRAFT v0 ready | `culturehub-la-2026/` |
+| CalArts CCAT — Technical Director, BB6 studio (job, not grant) | rolling (listed start 2026-04-13, past) | $100k–$125k / yr | DRAFT v0 ready | `calarts-ccat-tech-director-2026/` |
+| ~~Francis Gallery solo painting exhibition~~ | parked | — | parked — public Summer Exhibition deadline was 2026-04-30 (now past); track 2027. If Fia's source is a different/private call, restart this folder. | `francis-gallery-2026/` |
+>>>>>>> ba94c7807 (grants/SCORE: sync deadlines table with current folder state)
| Smart History — Contributor essay | rolling | unpaid | PITCH.md drafted (send first; PhD-or-equiv clause) | `smarthistory-essay/` |
### Drafted vs. submitted
@@ -25,23 +31,27 @@ draft to fit, fill in unanswered questions, and submit.
## Sourced — researching (not yet drafted)
-Surfaced in the 2026-05-19 Fia sweep (iMessage + email). Public-OK
-calls; sensitive sourcing context stays in the vault honeydo.
+From the 2026-05-19 Fia sweep (iMessage + email). See the dated block
+in `personal/honeydo.md` for the source quotes.
| Opportunity | Deadline | Amount | Fit / next |
|---|---|---|---|
-| SSRC Just Tech Fellowship | 2026-06-28 (23:59 EST) | up to $60,000 / yr | Tech-and-society creative practice — strong AC fit. Promote to `grants/ssrc-just-tech-2026/` + RESEARCH. |
-| Pollock-Krasner Foundation | rolling | up to $50,000 | Visual artists w/ exhibition record + financial need; Jeffrey qualifies (KADIST/SMK). Could fund SF / PAX East travel. |
-| Critical Playground (pitch) | unconfirmed | unknown | criticalplayground.org publication; Smarthistory-style pitch-first. Verify the call exists before drafting. |
-| CAC Individual Artist Fellowship — **Legacy** (via LAPP, LA County) | 2026-06-06 (23:59 PT) | $50,000 unrestricted | DRAFT v0 ready in vault `cac-iaf-2026/`. Statewide California Arts Council program, regional admin = Los Angeles Performance Practice. 5 narrative answers + samples plan + letter-of-support plan; canonical 2026 guideline PDF cached locally. |
+| SSRC Just Tech Fellowship | 2026-06-28 (23:59 EST) | up to $60,000 / yr | Tech-and-society creative practice — strong AC fit. Promote to `grants/ssrc-just-tech-2026/` + RESEARCH. Source: Fia link 2026-05-14. |
+| Pollock-Krasner Foundation | rolling | up to $50,000 | Visual artists w/ exhibition record + financial need; Jeffrey qualifies (KADIST/SMK). Could fund SF / PAX East travel. Source: Fia 2026-05-15 ("got it easily before"). |
+| Critical Playground (pitch) | unconfirmed | unknown | criticalplayground.org publication; Smarthistory-style pitch-first. Verify the call exists. Source: Fia 2026-04-15. |
+| CAC Individual Artist Fellowship — **Legacy** (via LAPP, LA County) | 2026-06-06 (23:59 PT) | $50,000 unrestricted | DRAFT v0 ready · 5 narrative answers (Legacy tier) + samples plan + letter-of-support plan + checklist in `cac-iaf-2026/DRAFT.md`. Local guideline PDF in `cac-iaf-2026/reference/`. Next: ask Casey Reas for the colleague letter (required, uploaded by him). |
## Submitted / Decided
+<<<<<<< HEAD
| Application | Submitted | Decided | Outcome | Folder |
|---|---|---|---|---|
| CultureHub — LA Residency 2026 | 2026-05-21 | — | **Submitted, awaiting decision** — *Notepat Jam*: a six-channel audio piece performed solo on six salvaged laptops booted into AC Native (each voicing one channel of the 5:1 Kalio surround), plus open programming (public workshop + dialogue). Answers on Desktop: `CultureHub-LA-Residency-2026-Application.md`. | `culturehub-2026/` |
| Tezos Foundation — Ecosystem Grant (KidLisp Keeps v12 + LA adoption) | 2026-05-22 | — | **Submitted, awaiting decision** — *KidLisp Keeps v12*: $45,000 across 5 milestones (~14 weeks). v12 contract (rotatable signer, role separation, holder-governed metadata, opt-in v11→v12 migration) released as a reusable Smart Contract Template, plus tutorial videos, custom keeps wallet polish, LA adoption sessions at NELA Computer Club, and an outreach campaign into fxhash/objkt. Rolling Q2 cycle → reviewed July 2026. Management: . | `tezos-foundation-keeps-2026/` |
| Are.na Annual Vol. 8 — pitch (theme "Score.") | 2026-04-20 | 2026-05-08 | **Rejected** — Meg Miller (editorial director). Channel `aesthetic-computer/self-teaching-scores` (68 blocks) + pitch page stay live; track for Vol. 9. | `gigs/are-na-annual-vol-8/` |
+=======
+(none yet — see `grants/lacma-2026/` for prior reference.)
+>>>>>>> ba94c7807 (grants/SCORE: sync deadlines table with current folder state)
## Personal Logistics (not grants but tracked here)
diff --git a/marketing/campaigns/collectors-altman-facetime/cover-prompt.txt b/marketing/campaigns/collectors-altman-facetime/cover-prompt.txt
new file mode 100644
--- /dev/null
+++ b/marketing/campaigns/collectors-altman-facetime/cover-prompt.txt
@@ -0,0 +1,131 @@
+A photograph of a colored-pencil and gouache drawing on warm cream
+paper, in tall vertical portrait orientation. Confident hatching and
+striping build tone. Tapered pencil edges. Optical mixing in patterned
+surfaces. Visible paper grain. Print-work sensibility. Diegetic light
+only — never overlay glow. Editorial peer-horizontality: both figures
+share the same eye-line, neither one centered as a hero. Candid
+iPhone-snapshot tone, deadpan everyday-help — NOT cinematic, NOT
+neon-noir, NOT a portrait studio. NEITHER figure looks at the camera
+or acknowledges the viewer in any way; both are fully absorbed in
+their own tasks. The drawing dissolves softly to pure cream paper at
+the edges with no hard frame and no app UI overlays of any kind.
+
+The scene: jeffrey's lived-in los angeles home office in late
+afternoon. A long warm-walnut desk against a tall west-facing window.
+A pale oak floor with a small worn persian rug underfoot. Warm
+late-afternoon golden side-light falling across the desk from the
+right. One small monstera leaf at the corner of the desk. A
+half-drunk mug of black coffee on a cork coaster. A yellow legal pad
+with a few hand-written notes pushed to the edge. At the back-right
+corner of the desk, a small beige tp-link wireless router with two
+stubby antennas and tiny green leds; a short butter-yellow cat6
+ethernet cable loops out of its back and dangles down behind the desk.
+
+Two walnut chairs side-by-side at the desk, both figures seated at
+the same eye-line, both facing the desk at a slight 3/4 angle to the
+camera.
+
+On the right of the frame: Sam Altman, recognizably him — mid-late
+30s, clean-shaven, slight build, soft wavy brown hair parted to one
+side, small attentive eyes, the gentle slightly-tense mouth from his
+press photos. He wears a plain navy crewneck t-shirt under a
+heathered light-grey patagonia-style fleece vest, dark grey
+straight-leg trousers, and plain white low-top sneakers. His
+expression is visibly frustrated and a little embarrassed: brows knit
+gently, mouth pressed into a thin almost-grimace, faint flush across
+the cheeks. His right hand rubs the back of his own neck; his left
+hand hovers indecisively just over the trackpad of his own laptop,
+fingers slightly curled, not quite touching. He is leaning forward
+into the laptop with a faint slump of defeat. His attention is locked
+entirely on the frozen FaceTime tile — he has no awareness of the
+camera or of jeffrey.
+
+Sam's laptop, open on the desk directly in front of him: a plain
+silver Apple MacBook Pro, 14-inch, matte aluminum, completely
+unstickered, no decorations of any kind. The lid is open AWAY from
+the camera so the screen faces the viewer at a slight 3/4 angle —
+the matte aluminum lid-back is NOT visible. Render the aluminum
+chassis in cool grey colored-pencil cross-hatching, never an
+airbrushed metallic gradient.
+
+ON Sam's MacBook Pro screen: a macOS FaceTime call window dominates
+the screen, drawn in the same colored-pencil hatching as everything
+else — never a glossy ui render. The remote video tile fills most of
+the window: a recognizably east-asian man in his early 40s, short
+black hair, slim wire-rimmed glasses, a plain charcoal collared shirt,
+sitting in front of a softly out-of-focus modern apartment interior
+with a tall paper-shade lamp behind him and a sliver of a balcony with
+hanging laundry — clearly a chinese city apartment in evening light,
+not a sterile office. He is FROZEN mid-expression: mouth slightly open
+in the middle of a word, one hand raised in a mid-gesture, the image
+visibly pixelated and slightly stretched with classic codec
+macroblock artifacts hatched in as small flat colored squares. A small
+"poor connection" indicator with half-filled signal bars sits in the
+upper-left of the call tile, and a small self-view thumbnail in the
+corner of the call shows Sam's own face, also slightly frozen. A
+faint macOS menubar runs along the top of the screen and a faint
+dock along the bottom — no readable third-party brand text. The
+screen is NOT a thumbnail of the surrounding room — only the FaceTime
+call.
+
+On the left of the frame, seated beside Sam at the same eye-line:
+the man in the reference photos — recognizably him, medium-length
+brown hair tucked behind one ear, soft red plastic glasses dangling
+on a thin cord around his neck (not currently on his face). He wears
+a faded printed crewneck t-shirt — a small hand-screened design across
+the chest in muted coral and butter-yellow that reads as a tiny
+hand-lettered piece-name word but is not legible as a real brand —
+under an open soft-grey knit cardigan with sleeves pushed to the
+forearms, wide-leg medium-cobalt linen trousers, and worn black canvas
+low-top sneakers with white rubber soles.
+
+In jeffrey's lap, balanced calmly across his knees: his own laptop —
+a chartreuse citrus-green Apple MacBook Neo, clamshell open, the
+regular silver apple logo just visible on the (mostly hidden) lid.
+No stickers, no butterfly, no cards on the lid, nothing in front of
+the laptop in the camera plane. The Neo's screen is angled toward
+jeffrey himself, so the camera mostly reads the foreshortened keyboard
+and a sliver of screen edge — we see the saturated green chassis
+clearly but the screen content is not legible. Both of jeffrey's
+hands rest calmly on the Neo's keyboard, fingers in fluid motion (a
+soft pencil-hatched suggestion of typing, not a frantic blur) — he is
+running a network diagnostic / speedtest from his own machine while
+Sam waits on the call. His expression is calm, patient, and quietly
+capable: a faint closed-mouth half-smile, eyes attentive on his own
+keyboard. He is NOT looking at Sam, at Sam's MacBook Pro, or at the
+camera.
+
+Read of the picture: Sam came over because his FaceTime to a colleague
+in china keeps dropping. He's on his own silver MacBook Pro at
+jeffrey's desk; jeffrey is sitting beside him on his chartreuse Neo,
+calmly running a network diagnostic from his side while Sam tries to
+keep the call going. They are peers at the same desk — neither savior
+nor supplicant. The quiet dignity is in jeffrey's composure, not in
+any hero pose. The joke reads on its own: the most powerful person in
+ai cannot make a video call work, and the artist next to him is just
+calmly running a ping.
+
+Palette: warm cream paper as the ground; warm afternoon golden side-
+light hatched into the marks; warm walnut desk; muted persian-rug
+reds and blues; cool grey colored-pencil hatching on Sam's silver
+MacBook Pro chassis; one saturated chartreuse-citrus pop on jeffrey's
+MacBook Neo (the brightest object in the frame); heathered grey on
+Sam's fleece vest; cobalt blue on jeffrey's trousers; one warm coral
+pop on jeffrey's red-glasses cord; soft butter-yellow on the cat6
+cable; muted matte black on the router and on jeffrey's sneakers.
+Hand-drawn, hatched, print-tech-aware, print-comic restraint.
+
+Avoid: photoreal automotive or product rendering; cinematic neon glow;
+any css-overlay light (the warmth comes from diegetic sun hatched into
+the marks); ANY app, social-media, or story ui chrome beyond the
+genuine macOS FaceTime window (no progress bars, no avatars, no
+handles, no story chrome, no icons — this is just a drawing on cream
+paper, NOT a screenshot of an app); any readable real brand text
+(apple logo is ok, third-party app names are not); any visible content
+on the Neo's screen — only the matte green back-edge and keys are
+seen; any other recursive screen-of-the-room reflection; cartoon
+exaggeration; smoothed-out skin; figures looking at or acknowledging
+the camera; hero-pose centering; stickers or butterflies on either
+laptop; anything occluding either laptop in the camera plane. Keep
+both figures at one matter-of-fact eye-line and let the joke read on
+its own.
diff --git a/reports/2026-04-24-mvis-s3-prospectus-resale.md b/reports/2026-04-24-mvis-s3-prospectus-resale.md
new file mode 100644
--- /dev/null
+++ b/reports/2026-04-24-mvis-s3-prospectus-resale.md
@@ -0,0 +1,102 @@
+# MVIS Form S-3 — Resale Registration of High Trail Convertible Note Shares
+
+**Filed:** April 24, 2026 (SEC Registration No. 333-)
+**Filer:** MicroVision, Inc. (Nasdaq: MVIS)
+**Source PDF:** `2026-04-24-s3-prospectus.pdf` (SEC accession 0001493152-26-018973)
+**Closing price the day before filing (Apr 23, 2026):** $0.63
+**Shares outstanding (Apr 20, 2026):** 326,982,727
+**Shares outstanding if all converted:** 388,298,697
+
+---
+
+## What this filing actually is
+
+This is **not a new capital raise.** It is a *resale shelf* — MVIS is fulfilling its
+contractual obligation to register the shares that sit *underneath* the convertible
+notes High Trail Special Situations LLC and High Trail II LLC (both Hudson Bay
+Capital affiliates, managed by Sander Gerber) received on **February 23, 2026**.
+
+The notes are already issued. This filing just lets High Trail freely resell the
+common stock once they convert — or once they receive shares in lieu of cash for
+the monthly partial redemptions.
+
+| Term | Detail |
+|---|---|
+| Aggregate principal | ~$43M (Exchanged Note + new Cash Note) |
+| Outstanding as of Apr 20, 2026 | $33.9M (excl. repayment premium) |
+| Maturity | March 1, 2028 |
+| Coupon | Zero |
+| Initial conversion price | **$0.8819/share** |
+| Share cap (pre-stockholder vote) | **61,315,970 shares** (Nasdaq 5635(d)(2)) |
+| Monthly partial redemptions | Begin Apr 1, 2026 → Mar 1, 2028 (holder's election) |
+| Forced conversion trigger | Stock closes > **$2.00** for 20 consecutive VWAP trading days |
+| Beneficial ownership cap | 4.99% (toggleable to 9.99% on 61 days' notice) |
+| Security | First-priority perfected lien on **all bank + securities accounts** |
+| Minimum liquidity covenant | Greater of **$21.5M** or **110% of outstanding principal** |
+| Cash burn covenant | Yes (referenced; details in 8-K Exhibits 10.1/10.2) |
+
+## Selling stockholders
+
+| Holder | Shares before | % | Shares offered | Shares after | % |
+|---|---|---|---|---|---|
+| High Trail (HTSS LLC + HTSS II LLC) | 17,173,390 | 4.99% | 61,315,970 | 5,750,225 | 1.73% |
+
+Address: c/o Hudson Bay Capital Management LP, 290 Harbor Drive, 3rd Floor,
+Stamford CT 06902. Sander Gerber is the managing member of Hudson Bay Capital
+GP LLC.
+
+## What it means
+
+### 1. Dilution overhang is now formal
+61.3M registered shares = **18.7% dilution** on the current 326.9M float. The
+market has been pricing this in since February, but with an active resale
+prospectus, every conversion can hit the tape immediately.
+
+### 2. Conversion price is *above* the market
+$0.8819 conversion vs. $0.63 spot (Apr 23). High Trail loses money converting
+today. So the practical near-term path is **monthly cash redemptions**, not
+share conversions — which drains MVIS's $74.8M cash directly.
+
+### 3. The $2.00 forced-conversion trigger is the bull case
+If MVIS rallies and stays above $2 for 20 trading days, MVIS can wipe out the
+notes by forcing conversion. **No cash leaves the building, debt disappears.**
+This is the cleanest exit for the company and aligns with retail bull theses.
+
+### 4. The covenant math is tight
+- Cash on hand (per Q4 2025 earnings): **$74.8M**
+- Minimum liquidity floor: $33.9M × 110% = **$37.3M** (or $21.5M, whichever is higher → **$37.3M**)
+- Headroom before covenant breach: ~$37M
+- 2026 projected burn: **$65–70M**
+
+So absent revenue acceleration or a stock recovery, MVIS has roughly **6–7
+months** of operating runway before tripping the covenant — at which point
+High Trail can foreclose on the bank/securities accounts.
+
+### 5. The 19.99% share cap is the next catalyst
+The 61,315,970 figure exists because Nasdaq Rule 5635(d)(2) caps issuance at
+19.99% of pre-deal shares without a stockholder vote. If MVIS calls a vote to
+lift the cap, it signals they expect more dilution is necessary. Watch the
+proxy filings.
+
+### 6. Hudson Bay Capital is a known short-friendly counterparty
+The prospectus dedicates an unusual amount of language to "short squeeze" risk
+— a tell that the underwriters know HBC's reputation in the MVIS retail
+community. HBC has historically hedged convertible positions with short
+exposure, profiting on the spread regardless of which way the stock moves.
+
+## Bottom line
+
+This filing makes the February deal **operational** — High Trail can now start
+liquidating into the market on conversion. It does not change the fundamentals
+established in February: MVIS bought time and integration capital at the cost
+of:
+
+- A senior lien on its cash
+- A real risk of forced asset foreclosure if cash drops below ~$37M
+- Up to 18.7% dilution if the stock rallies above $0.8819
+- Debt elimination if the stock can hold above $2.00 for 20 days
+
+For the AIM-takeover thesis: this **does not preclude** a takeover, but a
+buyer would have to deal with the High Trail lien, which sits ahead of common
+equity. A change-of-control clause likely exists in Exhibits 10.1/10.2 (8-K
+Feb 24, 2026) — worth pulling those to confirm.
diff --git a/reports/2026-05-07-user-comparison.html b/reports/2026-05-07-user-comparison.html
new file mode 100644
--- /dev/null
+++ b/reports/2026-05-07-user-comparison.html
@@ -0,0 +1,110 @@
+
+
+
+
+ user funnel — ac vs sotce — 2026-05-07 23:54:13 UTC
+
+
+
+
+
user funnel — aesthetic.computer vs sotce.net
+
generated 2026-05-07 23:54:13 UTC
+
+
+
22,774combined signups
+
2,869combined handles
+
132sotce subscribers
+
+
+
+
+
+
+
aesthetic.computer
+
sotce.net
+
notes
+
+
+
+
+
Tenant first activity
+
2023-02-16
+
2024-06-17
+
First day with stats reported by Auth0 (proxy for tenant launch)
+
+
+
Auth0 signups (all-time)
+
18,415
+
4,359
+
Sum of /api/v2/stats/daily.signups (the /users endpoint caps total at 1k)
+
+
+
Auth0 logins (all-time)
+
27,768
+
8,593
+
Lifetime authentications
+
+
+
Active subscribers
+
—
+
132
+
Sotce is paid; AC is free. Source: sotce.net/sotce-net/subscribers
+
+
+
Handles claimed
+
2,714
+
155
+
MongoDB @handles split by 'sotce-' _id prefix
+
+
+
Handle adoption
+
14.7%
+
3.6%
+
Handles ÷ signups
+
+
+
Subscription rate
+
—
+
3.0%
+
Subscribers ÷ signups
+
+
+
+
+
+
+
diff --git a/reports/2026-05-18-crypto-portfolio-market-report.md b/reports/2026-05-18-crypto-portfolio-market-report.md
new file mode 100644
--- /dev/null
+++ b/reports/2026-05-18-crypto-portfolio-market-report.md
@@ -0,0 +1,130 @@
+# crypto portfolio + market report — 2026-05-18
+
+> Read-only audit. Public addresses only — no keys decrypted, no mnemonics
+> read, nothing signed or sent. Supersedes `2026-03-30-crypto-portfolio-market-report.md`.
+> Reproducible: `node banking/crypto/audit-readonly.mjs` (snapshot in
+> `banking/crypto/data/audit-2026-05-18.json`).
+
+## the blunt headline
+
+**Total liquid across all chains ≈ $4.70.** There is no trading / yield /
+"make cash right now" play here — there is no capital to deploy. The only
+genuinely immediate cash is **withdrawing your own ~7.5 tez sitting in the
+keeps contract (~$2.53)**. Everything past that is rebuilding the keeps
+creative-economy engine, which pays out over weeks via royalties + new
+collectors — a real business lever, not instant cash.
+
+Live prices used: tez **$0.3370**, ETH **$2085.42**, SOL **$83.83** (CoinGecko, 2026-05-18).
+
+## wallet balances
+
+### tezos
+| wallet | address | balance | usd |
+|---|---|---|---|
+| aesthetic.tez | tz1gkf8EexComFBJvjtT1zdsisdah791KwBE | 2.5640 tez | $0.86 |
+| kidlisp/treasury | tz1fEjGQrEE2LXNKqcpTYAJV16pbbFpLeNyd | 0.2843 tez | $0.10 |
+| keeps admin / royalty rx | tz1Lc2DzTjDPyWFj1iuAVGGZWNjK67Wun2dC | 0.5017 tez | $0.17 |
+| staging | tz1dfoQDuxjwSgxdqJnisyKUxDHweade4Gzt | 0.1239 tez | $0.04 |
+| **liquid subtotal** | | **3.474 tez** | **$1.17** |
+| keeps KT1 contract (withdrawable by admin) | KT1Q1irsjSZ7EfUN4qHzAB2t7xLBPsAWYwBB | 7.500 tez | $2.53 |
+| **tezos total** | | **10.97 tez** | **$3.70** |
+
+### ethereum (mainnet)
+| wallet | address | balance | usd |
+|---|---|---|---|
+| 4esthetic.eth | 0x5e6758C96A4cB5E2A1FE2E2772020dc8ad753b08 | 0.000342 ETH | $0.71 |
+| whistlegraph.eth | 0x238c9c645c6EE83d4323A2449C706940321a0cBf | 0.0000674 ETH | $0.14 |
+| eth-2 (citizen idx1) | 0xF637C4E11072C46A3bF7cF70ef32470562B5a479 | 0.0000035 ETH | $0.01 |
+| eth-unknown | 0x98eAc86755792e03D0f027cA8CcFa83818B994c4 | 0.0000023 ETH | ~$0 |
+| **ethereum total** | | **~0.000415 ETH** | **~$0.87** |
+
+### base l2
+All wallets ~0 (4esthetic dust 3.85e-7 ETH). No change since March.
+
+### solana
+| wallet | address | balance | usd |
+|---|---|---|---|
+| phantom | D5tLrs4Ubh3tcHxhSmorPmGrDsozuSLhQDGVoBbR5P9d | 0.001569 SOL | $0.13 |
+| axio trade | 6JRph4ZZf5jt17CZhyBM9SYVexLrKmyPVMGCTMA4A7Ps | 0 SOL | $0 |
+| **solana total** | | | **~$0.13** |
+
+`axio` holds 4 SPL token accounts — all pump.fun mints
+(`…pump`), all with **no liquidity pool / dead market ≈ $0**. Leftover dust
+from old pump.fun trades that rugged. Not a windfall. (Optional: closing the
+token accounts reclaims ~0.008 SOL rent — not worth the effort.)
+
+### bitcoin + cardano — AUDIT GAP
+4 BTC wallets + 1 Cardano wallet have **no public address stored in
+`wallets/wallets.json`** (`keys_in` points at ordinals.com / Sparrow). Cannot
+be audited read-only without them. **To include them next run, add a public
+`address` (or xpub / receive address) field for each — no private key
+needed.** Given they're all "ordinals-test"-type wallets, expected value is
+low, but unknown until addresses are supplied.
+
+## keeps NFT engine (KT1Q1irsjSZ7EfUN4qHzAB2t7xLBPsAWYwBB)
+
+- **88 tokens minted** (was 81 on 2026-03-30, +7). Each is 1/1.
+ *Caveat: TzKT "holder" count ≈ token-balance rows, not 88 distinct people —
+ do not quote "88 unique collectors."*
+- **Contract holds 7.5 tez** of accumulated mint proceeds/fees — withdrawable
+ by the admin wallet. **This is the single most concrete cash item.**
+- **We currently hold 4 tokens:** $odj #28, $pie #81, $tam #86, $r94 #87.
+- **Our live OBJKT listings (6):** $9x9 #76 @ 7 tez · $tam #86 @ 10 · $weus
+ #27 @ 12 · $ixi #48 @ 12 · $pie #81 @ 12 · $r94 #87 @ 14. (#28 $odj is held
+ but **unlisted**.)
+- **Standing offers on the collection (held by others, not us):**
+ - #3 $3jj — **100 tez** bid from `tz1QRXx…`
+ - #53 $fsf — **16.9 tez** bid from same `tz1QRXx…`
+ - These same two bids existed on 2026-03-30 — **7 weeks stale, holders
+ haven't sold.** If either sells, your royalty (10%) = **10 tez + 1.69 tez
+ ≈ 11.69 tez (~$3.94 today)**. Not in your control.
+- Royalty inflow is a trickle (0.05 / 0.08 / 0.11 tez receipts). Last contract
+ activity **2026-04-27** — collection has been quiet ~3 weeks.
+
+## delta vs 2026-03-30
+
+- **tez price collapsed ~$0.80 → $0.337 (−58%)** — the dominant reason the
+ position shrank in USD terms.
+- Liquid tez 7.17 → 3.47 (spent on 3× `keep` mints @ 2.5 tez on 04-23; 7.5 tez
+ now parked in the contract).
+- ETH balance actually rose 0.00015 → 0.000415, still dust at today's price.
+- Keeps minted 81 → 88; engine otherwise idle since late April.
+- The two big standing bids did **not** convert in 7 weeks.
+
+## ranked cash actions (honest)
+
+1. **Withdraw the 7.5 tez from the keeps KT1 contract** — already yours, one
+ admin call (`withdraw_fees`) from the keeps-admin wallet. ~$2.53. *You
+ execute it — this audit is read-only and never touches keys.*
+2. **List $odj #28** on OBJKT (held, currently unlisted) and sanity-check the
+ 6 live listings against comparable recent sales. A 10-tez sale ≈ $3.37.
+ Buyer-dependent.
+3. **Nudge the holders of #3 ($3jj) and #53 ($fsf)** that there's a 7-week
+ standing bid (100 / 16.9 tez) from `tz1QRXx…`. If they take it you collect
+ ~11.69 tez royalty. Indirect, but a stale bid worth surfacing.
+4. **Cross-chain distribution (strategic, NOT cash-now):** Zora open editions
+ on Base, Farcaster frames, Solana cNFTs. Near-zero cost, volume/visibility
+ — payoff is weeks out, not today.
+5. **Leave the dust.** ETH+SOL ≈ $1; conversion/gas costs exceed value. Don't
+ move it.
+6. **Ignore the pump.fun tokens** — dead, ~$0.
+
+## the realistic takeaway
+
+The "make cash right now" answer is: **~$2.50 by withdrawing your own
+contract balance, plus whatever the listings/standing-bids convert to over
+the coming weeks.** The keeps engine is the only real revenue lever and it's
+been idle ~3 weeks — re-activating it (new mints, driving secondary volume so
+royalties flow, cross-chain reach) is a multi-week creative-economy play, not
+a same-day cash event. Anyone framing a sub-$10 multi-chain position as a
+trading opportunity is wrong.
+
+## audit gaps / to improve next run
+
+- Add public BTC + Cardano addresses to `wallets/wallets.json` (no keys).
+- ETH NFT enumeration: keyless Reservoir is now API-gated. The vault already
+ has a **read-only** Etherscan/OpenSea data key (separate from wallet
+ secrets) — wiring that into `audit-readonly.mjs` would restore the NFT
+ inventory. Prior manual inventory still stands (4esthetic.eth: a2p-v1 ×3,
+ the-longest-whistlegraph-ever 4/5, screenshots-14 ×2; whistlegraph.eth:
+ collected works) — all illiquid art, not cash.
diff --git a/reports/2026-05-18-keeps-opportunistic-targeting.md b/reports/2026-05-18-keeps-opportunistic-targeting.md
new file mode 100644
--- /dev/null
+++ b/reports/2026-05-18-keeps-opportunistic-targeting.md
@@ -0,0 +1,79 @@
+# keeps opportunistic targeting — runbook (2026-05-18)
+
+> DRAFT for @jeffrey to execute. This audit is read-only: no keys decrypted,
+> nothing signed or sent by the agent. Every on-chain step below is yours to
+> run/sign. Built from `2026-05-18-nft-attention-map.md`.
+
+## honest expected outcome
+
+~**$50–150 over the coming weeks**, from targeted airdrops + royalties on
+resale — not "cash right now." The real value is **re-activating the keeps
+engine** (idle ~3 weeks) by putting tokens in front of proven active
+collectors. Median Tezos NFT sale is ~$3; this only works because it's
+*targeted at specific funded buyers*, not broadcast.
+
+## priority targets (verified, this week, public data)
+
+| # | wallet | identity | on-hand | why |
+|---|---|---|---|---|
+| 1 | `tz1WrEjvFiVTChGdxAfH8qyB3v9KyN97uUTg` | **Louis** (OBJKT) | **753 tz (~$254)** | March's #1 keeps-airdrop recipient ($b00, 5,500 tz/14d). Still buying this week. Funded. Already engaged once → warm. |
+| 2 | `tz1VwAPwtHtQ2jADWSzBJ8kBQQMV2NtcU4W8` | **Jose Antonio Ojeda** · `ojedasbodega.tez` | 531 tz (~$179) | Collector since 2021 (15.8k tx), 7 buys/3 collections this week, funded. |
+| 3 | `tz1c2ayR85dUMXZAvNixTJz1WdW9TqvUgfRq` | **6ENERATOR** · `generator.tez` | 1.8 tz | Tastemaker since 2021 (14.9k tx). Spends fast (low idle balance), 9 buys/6 collections this week. Distribution/visibility value. |
+| 4 | `tz2DviPC4iRCWBvTktAPLxauX5W6URa6WPY2` | (anon) | 131 tz | Newer (Jan 2026) but diversified — 5 buys/5 collections. Secondary. |
+| — | `tz2JpM2Z7p1iWLY5DgBa23xLU1RY1tqhPHQZ` | (anon) | 0.6 tz | Was a 514-tz single-collection whale; balance now near-zero (moved funds). **Skip unless it re-funds.** |
+
+## the asset to deploy
+
+You hold 4 keeps free to use: **$odj #28 (unlisted)**, $pie #81, $tam #86,
+$r94 #87. Two routes:
+
+- **A — airdrop existing held tokens.** Zero mint cost. Send $odj #28 (your
+ only unlisted held token) to target #1, and re-home one of #81/#86/#87.
+ Fastest, no new work.
+- **B — mint a fresh keep into the glitch lane (highest-$ shot).** A whale
+ swept ~7,150 tz of `HYPER[]GLITCH` this week; the glitch/visual-corruption
+ aesthetic is hot and kidlisp does it natively. Mint one corruption-themed
+ keep (~2.5 tz cost), list it ~150–250 tz, and offer/airdrop a copy toward
+ the glitch buyers + target #1. Best single-sale upside (~$50–90 if it
+ lands).
+
+Recommended: **A for targets #1–#3 now** (free, re-activates engine
+immediately) + **B as the one revenue swing** this week.
+
+## runbook (you execute — keys never leave your hands)
+
+1. **Withdraw the 7.5 tz already in the contract first.** From the keeps-admin
+ wallet, call `withdraw_fees` on `KT1Q1irsjSZ7EfUN4qHzAB2t7xLBPsAWYwBB`.
+ That's the only genuine cash-now (~$2.50) and funds the mint below.
+2. **(Route A) Airdrop held tokens.** FA2 `transfer` on the keeps contract:
+ `$odj #28 → Louis (tz1WrEjv…)`; one of $pie/$tam/$r94 → `ojedasbodega.tez`.
+ (Tokens currently listed must be de-listed on OBJKT first or the transfer
+ will conflict with the active ask.)
+3. **(Route B) Mint the glitch keep.** `keep` entrypoint (~2.5 tz), kidlisp
+ corruption piece, IPFS via the self-hosted lith node. List ~150–250 tz on
+ OBJKT. Place it on the radar of the glitch buyers (offer or airdrop a
+ variant to target #1).
+4. **Surface the stale standing bids.** `KT1…` has open offers of **100 tz on
+ #3 ($3jj)** and **16.9 tz on #53 ($fsf)** from `tz1QRXx…`, untouched 7
+ weeks. You don't hold #3/#53, but a public nudge to those holders (DM /
+ chat invite) that there's a live 100-tz bid → if they sell, you collect
+ **10% royalty (≈11.7 tz / ~$3.9)**. Free to do.
+5. **Re-list $odj #28** if you keep it instead of airdropping (held, currently
+ unlisted) and sanity-check the 6 live listings (7–14 tz) vs comparable
+ recent sales (~8.8 tz median — your floor listings may be over-priced for
+ the current market).
+
+## still blocked: ETH/Base extension
+
+You authorized the read-only OpenSea/Zora data key, but the harness safety
+layer independently blocks the agent from scanning vault credential files —
+even with your approval — and I won't route around that. To unblock, pick one:
+
+- **Tell me the exact location/var name** (e.g. "OpenSea key is
+ `OPENSEA_API_KEY` in `vault//.env`") so I read only that, no scanning.
+- **Add a Bash permission rule** allowing the read, then I retry.
+- **Paste it** into a throwaway env var yourself: `set -x OPENSEA_API_KEY …`
+ in a `!`-prefixed prompt, and I'll use `$OPENSEA_API_KEY` without ever
+ reading the file.
+- **Stay Tezos-only** — recommended; ETH-side has the same ~$3–90 ceiling and
+ needs new infra, while everything above is ready to run today.
diff --git a/reports/2026-05-18-nft-attention-map.md b/reports/2026-05-18-nft-attention-map.md
new file mode 100644
--- /dev/null
+++ b/reports/2026-05-18-nft-attention-map.md
@@ -0,0 +1,97 @@
+# NFT attention map — 2026-05-18
+
+> Read-only network survey. Source: OBJKT GraphQL (keyless) + fxhash, last 48h
+> of on-chain sale events (500 events, price > 0). ETH/Base: see gap below.
+> Companion to `2026-05-18-crypto-portfolio-market-report.md`.
+
+## the honest ceiling first
+
+Tezos NFT sale price distribution, last 48h (500 sales):
+
+| min | median | p90 | p99 | max |
+|---|---|---|---|---|
+| 0.05 tz | **8.8 tz (~$3)** | 275 tz (~$93) | 30,000 tz | 10,000,000 tz |
+
+The p99/max are **wash trading** (`hic et nunc` 10M tz/24 "sales",
+`_syntax` 69,420 tz, `EmProps` 10k-tz medians with 2 addresses). Filtered out
+below. The **real** market is single-digit-to-low-hundreds of tez. A median
+sale is **~$3**. A genuinely "hot" sale is ~$90. "Some cash" on this surface
+realistically means **tens of dollars from a well-targeted drop**, not an
+income event. Mapping attention here is worth doing — but size expectations
+to the data.
+
+## where the real attention is (wash-filtered, ranked by distinct buyers)
+
+| collection | distinct buyers | sales | median | read |
+|---|---|---|---|---|
+| **open objkt** (OE / minting factory) | 12 | 26 | 8.8 tz | broadest genuine attention — cheap open-edition mint surface |
+| hic et nunc / teia | 8 | 23 | 120 tz | OG community still transacting at real prices (~$40) |
+| Friedeberg | 2 | 9 | 23 tz | small, concentrated |
+| CONTINGENT | 2 | 10 | 100 tz | whale-concentrated |
+| HYPER[]GLITCH | 1 | 41 | **180 tz** | **one whale swept ~7,150 tz of glitch art** |
+| TAPE LOOP CITY | 1 | 45 | 35 tz | one collector sweeping a cheap series |
+| fx(hash): Genesis | 1 | 40 | 4 tz | one collector sweeping cheap fxhash |
+
+**Pattern:** attention at this scale is **a person, not a category**. The
+"trending" tail is individual collectors sweeping one series each. The only
+broad-distribution surface is cheap OBJKT open editions (~$3/sale).
+
+## the attention graph — who is spending right now
+
+| address | buys (48h) | tz | breadth | note |
+|---|---|---|---|---|
+| `tz1c2ayR85dUMXZAvNixTJz1WdW9TqvUgfRq` | 9 | 123 | **6 collections** | most active diversified collector — prime target |
+| `tz1VwAPwtHtQ2jADWSzBJ8kBQQMV2NtcU4W8` | 7 | 96 | 3 | solid active collector |
+| `tz2JpM2Z7p1iWLY5DgBa23xLU1RY1tqhPHQZ` | 5 | 514 | 1 | single-collection whale (glitch/EmProps lane) |
+| `tz2DviPC4iRCWBvTktAPLxauX5W6URa6WPY2` | 5 | 40 | 5 | diversified, low-budget |
+| **`tz1WrEjvFiVTChGdxAfH8qyB3v9KyN97uUTg`** | 3 | 48 | 3 | **= March's #1 keeps-airdrop recipient ($b00, 5,500 tz/14d). Still active 7 weeks later.** |
+| `tz1aYMJM8ke8msTo2LyGwdMBxxdbFsKmfBDK` | 25 | 4 | 1 | mint-farmer (free/cheap OE spam) — ignore |
+
+## the opportunistic play (concrete, honest)
+
+Broad minting is **not** the move — median $3, and the "trending" collections
+are single whales, not crowds. The only thing that has ever moved keeps is the
+**March playbook: mint → airdrop to proven active spenders → royalty on their
+resale.** It still applies, and the targets are now identified:
+
+1. **Re-engage `tz1WrEjv…`** — March's top airdrop recipient, *still buying
+ today*. Highest-confidence single contact: mint 1 keep, airdrop to this
+ exact address. Proven 5,500 tz spender who already engaged once.
+2. **Airdrop the diversified actives** — `tz1c2ayR…` (6 collections),
+ `tz1VwAPw…`, `tz2DviPC…`. Real taste, buying this week, low spam risk.
+3. **One glitch-themed keep into the whale lane** — a whale swept ~7,150 tz of
+ HYPER[]GLITCH; another (`tz2JpM2Z…`) spent 514 tz single-collection.
+ kidlisp does visual-corruption aesthetics natively. A glitch-themed 1/1
+ keep, listed ~150–250 tz and offered directly to those wallets, is the
+ single highest-$ shot (~$50–90 if it lands).
+4. Funnel only: a cheap open edition rides the "open objkt" 12-buyer surface
+ for visibility — but that's ~$3/sale, count it as marketing not revenue.
+
+**Realistic upside of the whole play: ~$50–150 over the coming weeks**, mostly
+from (1)+(3), driven by targeted airdrops + royalties, not volume. Plus it
+re-activates the keeps engine (idle ~3 weeks).
+
+## ETH / Base — AUDIT GAP (needs a decision)
+
+Every keyless ETH/Base NFT trending source is dead or gated:
+- OpenSea v2 → 401 (key required)
+- Reservoir → key-gated (collections + users)
+- Zora discover/trending → 404 (public API moved/gated)
+- CoinGecko keyless `/nfts/list` → returns only static blue-chips
+ (Punks/BAYC/Pudgy/Azuki) with no real volume ordering — useless for
+ *new/opportunistic* signal, and those aren't mint opportunities anyway.
+
+The real ETH-side opportunistic-mint surface in 2026 is **Zora/Base open
+editions + Farcaster mints**, which requires a data API key. The vault holds a
+**read-only** OpenSea data key (distinct from any wallet secret). I did **not**
+reach for it — using a vault credential is your call. Options:
+- **Authorize the read-only OpenSea/Zora data key** → I extend this map to
+ Base/Zora/Farcaster mint activity.
+- **Skip ETH** → focus the (cheap, fast, fully-mapped) opportunistic play on
+ Tezos, where the keeps infra already exists.
+
+## reproduce
+`/tmp/tez_events.json` is the raw 48h pull. Survey logic is ad-hoc jq over the
+OBJKT GraphQL `event`/`fa` types (schema notes: `event.price_xtz`,
+`event.recipient_address` = buyer, `fa.volume_24h` is stale — aggregate from
+`event` instead).