feat(account): portfolio composition report

[piekstra]

What

New read-only command lofty account portfolio (--json → account-portfolio/v1). It answers "where is my money, and what pays rent?":

Weighting	Basis
by state	share of holdings' estimated value
by city (city + state)	share of holdings' estimated value
by property	share of holdings' estimated value
by rent	each property's share of projected daily rent; non-paying properties excluded (null share)
renting vs non-renting	share of properties renting, and share of holdings value that is rent-earning

How the numbers are derived

• Value: each position's currentValue from /account/positions (tokens × current price; verified equal on live data). Fully sold positions (0 tokens) are skipped.
• Daily rent: listing projected_annual_cash_flow ÷ issued tokens × tokens held ÷ 365. monthly_rent is not used because it reads 0 on occupied, distributing properties.
• Renting: true unless the listing has is_occupied: false (vacant), is_delinquent: true, or projects no cash flow. rentStatus reports which.
• A held property whose listing returns 404 is still weighted by value under an unknown location (with a stderr note), so the other shares are not inflated.

Inputs: one /account/positions call plus one /properties/{id} per held property. No writes.

Tests

• 9 unit tests in src/commands/portfolio.rs over new synthetic fixtures (tests/fixtures/account/positions-portfolio.json, tests/fixtures/portfolio/property-listings.json): state/city/property weights, rent-share exclusion, renting share by count and value, missing listing, zero cash flow, empty portfolio.
• Shape test pinning the listing fields the report reads (city, state, tokens, projected_annual_cash_flow, is_occupied, is_delinquent) against the captured property-get.json.
• Surface test: account portfolio --help renders.
• make verify green locally. Run once against the live API with the installed key: exit 0, schema and field types as documented.

[piekstra-dev]

Automated PR Review

Reviewed commit: fc8d1acddaa2
Profile: reviewer - Posting as: piekstra-dev

Summary

Reviewer	Findings
rust:implementation	0
rust:implementation-tests	0
policies:conventions	0
architecture:solid	2

architecture:solid (2 findings)

Minor - src/commands/account.rs:152

U-S2: the Portfolio arm repeats the core's rule for what counts as a holding (currentTokens > 0, a present propertyId) to decide which listings to fetch, and portfolio::portfolio applies that rule again on its own (portfolio.rs:78-86). The two copies already differ. The shell skips a position with no propertyId, but the core keeps it as a row with id "" under unknown. A later change to either rule (for example, treating dust balances as sold) will fetch one set of listings and weight a different set of rows. Suggested fix: expose pub fn held_property_ids(positions: &[Value]) -> Vec<&str> (or a held_positions iterator) from portfolio.rs. Use it in both the fetch loop and portfolio(), so the shell only translates input and calls the core, and the holding rule lives in one tested place.

Minor - src/commands/portfolio.rs:20

U-O1 / U-L2: rent status is a closed domain set (renting, vacant, delinquent, no-cash-flow, unknown) carried as &'static str, and behaviour branches on string equality in three separate places: the daily-rent guard (line 94), the renting filter (line 113), and the per-property renting/rentSharePct decision (line 154). A typo or rename in any one comparison compiles cleanly and silently zeroes rent or miscounts renting properties. Idiomatic Rust gives this an exhaustive type. Suggested fix: enum RentStatus { Renting, Vacant, Delinquent, NoCashFlow, Unknown } with fn is_renting(self) -> bool and fn as_str(self) -> &'static str (or #[derive(Serialize)] #[serde(rename_all = "kebab-case")]). Store it in Row and serialize only at the JSON boundary, so the account-portfolio/v1 output stays the same. Low blast radius because the type is module-private, so not blocking.

Reviewer Coverage

• rust:implementation — complete (broad); inspected 3 assigned files (10 inspected across reviewers): src/commands/account.rs, src/commands/mod.rs, src/commands/portfolio.rs; skipped: none; constraints: none
• rust:implementation-tests — complete (broad); inspected 5 assigned files (10 inspected across reviewers): src/commands/portfolio.rs, tests/cli_surface.rs, tests/fixture_shapes.rs, tests/fixtures/account/positions-portfolio.json, tests/fixtures/portfolio/property-listings.json; skipped: none; constraints: none
• policies:conventions — complete (broad); inspected 5 assigned files (10 inspected across reviewers): AGENTS.md, README.md, src/commands/account.rs, src/commands/mod.rs, src/commands/portfolio.rs; skipped: none; constraints: No sibling cli-common/.github checkouts were present in this workbench, so shared Open CLI Collective docs could not be checked; relied on repo-local AGENTS.md/README.md as the only available convention source.
• architecture:solid — complete (broad); inspected 4 assigned files (10 inspected across reviewers): src/commands/account.rs, src/commands/mod.rs, src/commands/portfolio.rs, tests/fixtures/README.md; skipped: none; constraints: Could not run make verify, clippy, or cargo test: shell commands in the checkout required interactive approval that this non-interactive session cannot grant. Build and test status is taken from the PR description, not reproduced. git diff was unavailable for the same reason; changed hunks were identified by reading head files against the change map and PR description.

Inspected files (10)

• AGENTS.md
• README.md
• src/commands/account.rs
• src/commands/mod.rs
• src/commands/portfolio.rs
• tests/cli_surface.rs
• tests/fixture_shapes.rs
• tests/fixtures/README.md
• tests/fixtures/account/positions-portfolio.json
• tests/fixtures/portfolio/property-listings.json

0 PR discussion threads considered. 0 summarized; 0 resolved.

---

Completed in 2m 19s | $2.35 | claude-sonnet-5, claude-opus-5-5 | cr 0.10.314

Field	Value
Model	claude-sonnet-5, claude-opus-5-5
Reviewers	rust:implementation, rust:implementation-tests, policies:conventions, architecture:solid
Engine	claude_cli · claude-sonnet-5, claude-opus-5-5
Reviewed by	cr · piekstra-dev
Duration	2m 19s wall · 5m 05s compute
Cost	$2.35
Tokens	92 in / 22.1k out

Per-workstream usage

• orchestrator-selection — claude-sonnet-5
  • In: 6
  • Out: 2.2k
  • Cache read: 139.9k
  • Cache create: 17.0k
  • Cost: $0.12
  • Duration: 28s
• rust:implementation — claude-sonnet-5
  • In: 12
  • Out: 3.3k
  • Cache read: 410.9k
  • Cache create: 46.4k
  • Cost: $0.30
  • Duration: 44s
• rust:implementation-tests — claude-sonnet-5
  • In: 24
  • Out: 4.6k
  • Cache read: 697.8k
  • Cache create: 33.7k
  • Cost: $0.32
  • Duration: 59s
• policies:conventions — claude-sonnet-5
  • In: 30
  • Out: 6.6k
  • Cache read: 1.0M
  • Cache create: 60.4k
  • Cost: $0.51
  • Duration: 1m 36s
• architecture:solid — claude-opus-5-5
  • In: 14
  • Out: 4.9k
  • Cache read: 363.4k
  • Cache create: 85.8k
  • Cost: $0.86
  • Duration: 1m 04s
• orchestrator-rollup — claude-sonnet-5
  • In: 6
  • Out: 457
  • Cache read: 158.0k
  • Cache create: 21.6k
  • Cost: $0.24
  • Duration: 11s

[piekstra-dev]

Automated PR Review

Reviewed commit: 3b2e33a1e2e1
Profile: reviewer - Posting as: piekstra-dev

Summary

Reviewer	Findings
rust:implementation	0
rust:implementation-tests	0
policies:conventions	0
architecture:solid	0

Reviewer Coverage

• rust:implementation — complete (constrained); inspected 3 assigned files (10 inspected across reviewers): src/commands/account.rs, src/commands/mod.rs, src/commands/portfolio.rs; skipped: none; constraints: none
• rust:implementation-tests — complete (constrained); inspected 5 assigned files (10 inspected across reviewers): src/commands/portfolio.rs, tests/cli_surface.rs, tests/fixture_shapes.rs, tests/fixtures/account/positions-portfolio.json, tests/fixtures/portfolio/property-listings.json; skipped: none; constraints: none
• policies:conventions — complete (constrained); inspected 5 assigned files (10 inspected across reviewers): AGENTS.md, README.md, src/commands/account.rs, src/commands/mod.rs, src/commands/portfolio.rs; skipped: none; constraints: No sibling cli-common/.github checkouts were present in this workbench, so shared Open CLI Collective docs could not be checked; relied on repo-local AGENTS.md/README.md as the only available convention source.
• architecture:solid — complete (constrained); inspected 4 assigned files (10 inspected across reviewers): src/commands/account.rs, src/commands/mod.rs, src/commands/portfolio.rs, tests/fixtures/README.md; skipped: none; constraints: Both settled threads were re-checked against head 3b2e33a. The RentStatus enum (portfolio.rs:15) and the shared held() helper (portfolio.rs:74, used at account.rs:151) resolve them, so neither is re-raised. Did not run make verify, clippy, or cargo test: shell commands in the checkout need interactive approval, which this session lacks. Build and test status comes from the PR description, not from a reproduction.

Inspected files (10)

• AGENTS.md
• README.md
• src/commands/account.rs
• src/commands/mod.rs
• src/commands/portfolio.rs
• tests/cli_surface.rs
• tests/fixture_shapes.rs
• tests/fixtures/README.md
• tests/fixtures/account/positions-portfolio.json
• tests/fixtures/portfolio/property-listings.json

0 PR discussion threads considered. 0 summarized; 0 resolved.

---

Completed in 56s | $3.30 | claude-sonnet-5, claude-opus-5-5 | cr 0.10.314

Field	Value
Model	claude-sonnet-5, claude-opus-5-5
Reviewers	rust:implementation, rust:implementation-tests, policies:conventions, architecture:solid
Engine	claude_cli · claude-sonnet-5, claude-opus-5-5
Reviewed by	cr · piekstra-dev
Duration	56s wall · 2m 19s compute
Cost	$3.30
Tokens	54 in / 8.4k out

Per-workstream usage

• rust:implementation — claude-sonnet-5
  • In: 8
  • Out: 1.1k
  • Cache read: 390.7k
  • Cache create: 19.4k
  • Cost: $0.47
  • Duration: 14s
• rust:implementation-tests — claude-sonnet-5
  • In: 14
  • Out: 2.4k
  • Cache read: 622.6k
  • Cache create: 20.4k
  • Cost: $0.55
  • Duration: 40s
• policies:conventions — claude-sonnet-5
  • In: 16
  • Out: 2.5k
  • Cache read: 932.1k
  • Cache create: 21.8k
  • Cost: $0.81
  • Duration: 38s
• architecture:solid — claude-opus-5-5
  • In: 10
  • Out: 2.0k
  • Cache read: 482.1k
  • Cache create: 22.9k
  • Cost: $1.18
  • Duration: 33s
• orchestrator-rollup — claude-sonnet-5
  • In: 6
  • Out: 371
  • Cache read: 187.7k
  • Cache create: 3.5k
  • Cost: $0.30
  • Duration: 12s