[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AP3K4J7JWZRWX62TYADZ4G":3,"skill-same:01M7AP3K4J7JWZRWX62TYADZ4G":64},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":33,"ownerDescription":34,"updatedAt":35,"install":36,"resources":49},"01M7AP3K4J7JWZRWX62TYADZ4G","alexandria",null,"Preserve heterogeneous lending-protocol captures by digest and expose a narrow, source-bound credit view for Tabularium and Probitas. Use when the user names Alexandria or asks to archive lending data for reproducible, address-scoped credit research. Raw release and registered Aave v4 and Clearpool derivation, disposable indexing, address queries and a checked-in offline demonstration, unsigned in-toto release statements, a bounded Compound v3 Phase 0 method proof and a resumable Ethereum USDC interval collector with per-subject implementation epochs and preserved Compound, Wildcat V1, Wildcat V2 and Aave V3 mainnet intervals are available.","wildcat-finance\u002Fskills","wildcat-finance","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F113041915?v=4","01M7AP0V1G8X0SWHP6C5V90RM5","The Wildcat Protocol","01M7AP0V24Y50BGVCQ4DV496VD",13,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Falexandria\u002Fskills\u002Falexandria\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: alexandria\ndescription: >\n  Preserve heterogeneous lending-protocol captures by digest and expose a\n  narrow, source-bound credit view for Tabularium and Probitas. Use when the\n  user names Alexandria or asks to archive lending data for reproducible,\n  address-scoped credit research. Raw release and registered Aave v4 and\n  Clearpool derivation, disposable indexing, address queries and a checked-in\n  offline demonstration, unsigned in-toto release statements, a bounded\n  Compound v3 Phase 0 method proof and a resumable Ethereum USDC interval\n  collector with per-subject implementation epochs and preserved Compound,\n  Wildcat V1, Wildcat V2 and Aave V3 mainnet intervals are available.\nmetadata:\n  version: \"3.13.0\"\n---\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"..\u002F..\u002Fassets\u002Fcharacters\u002Falexandria.png\" width=\"1200\">\n\u003C\u002Fp>\n\n# Alexandria\n\n## Frontier\n\nAlexandria owns its own preservation and credit-view frontier, not Hexaemeron's delivery or\nSolidity frontier. Its version, held target, next job, and maturity\nstate live in [EVOLUTION.md](EVOLUTION.md). Do not recommend or run\nanother frontier pass after that ledger becomes mature.\n\n\u003C!-- marketplace-context:start -->\n## Where this sits\n\nAlexandria preserves heterogeneous lending captures byte for byte, then exposes only the source-bound credit view a reviewed mapping can defend.\n\n**Current frontier.** Ordinary builds now emit `alexandria-interval-receipt\u002Fv2`, which attributes each preserved proxy log to an implementation epoch by block, transaction index and log index, and `check` re-derives every owner offline; reconciliation still compares a log without its transaction index, so a second provider that reports a different index for the same log records `agreed`.\n\u003C!-- marketplace-context:end -->\n\nAlexandria is the archive and catalogue behind durable lending-protocol\nresearch. Raw captures remain unchanged. Tabularium owns their interpretation\nas venue-qualified credit events, and Probitas may consume those events in a\ndossier. Ariadne may bind a finished release to its evidence. Lazarus is the\nneighbouring specialist for the finite historical Ethereum state and exact RPC\ntraffic one application test needs; it does not replace Alexandria's dataset\nboundary.\n\nSynkrisis compares validated agent-run observations through a checked cohort,\nbounded findings, a fixed report, and whole-path verification. It cannot\nreinterpret an Alexandria release, treat release rows as run observations, or\nauthorise a new capture.\n\n`$SKILL_DIR` is the directory holding this file. The command lives at\n`$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py`; resolve it from where you loaded this\nskill.\n\n## Day to day\n\n**Research.** A protocol endpoint or hosted indexer may disappear. Preserve\nthe exact response, its capture scope and its gaps before the only cheap copy\nis gone.\n\n**Credit.** A counterparty record needs events, position observations and an\naccount of the venues and intervals checked. An empty result without coverage\nis not a clean history.\n\n**Data engineering.** Raw GraphQL responses, chain logs and replay fixtures\ndo not need one payload schema. Bind their bytes in one manifest, then build a\nsmall, versioned credit view that points back to them.\n\n## Raw releases\n\nPrepare a capture plan using the contracts under\n`$SKILL_DIR\u002F..\u002F..\u002Fschemas\u002F`. Every component path is relative to the plan's\ndirectory. Every component has a `public`, `restricted` or `private` access\nclass and a `permitted`, `restricted`, `prohibited` or `unknown` redistribution\nclass. Give every capture a venue, an `eip155` chain, a non-secret source\nreference, an evidence and finality class, a full-dataset or subject-scoped\nboundary and counted coverage or a stated gap. Then ingest it:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" ingest \\\n  --plan capture-plan.json --output release\n```\n\nIngest reads only the plan and its confined local source files. It copies raw\nbytes unchanged into digest-derived paths, writes through a temporary sibling\ndirectory and installs the release atomically. It refuses absolute paths,\ntraversal, symlinks and replacement of a different release. Record the\n`sha256:...` release ID printed on success.\n\nVerify before using or moving a release:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" verify release\n```\n\nVerification is offline and read-only. It checks canonical manifest bytes,\nrelease identity, object paths, byte counts, digests, component access and\nredistribution classes, capture source, scope, evidence and finality classes,\ncollection counts, declared gaps, correction links and exact release-tree\nmembership. It does not establish publisher identity, source completeness or\nchain finality.\n\nA `proof-backed-state` capture earns that class only during `verify`: its\ncomponent must be a Lazarus manifest, every named fixture file must be present\nby digest and byte count, Alexandria must reconstruct and rerun Lazarus, and\nthe capture must stay within the proved block and targets. Finality must be\n`unknown` because Lazarus proves block binding but reports no finality class.\nThe scope must be subject-scoped and remain within the proof targets because a\nfinite proof set is not a full dataset. Missing Lazarus or its pinned packages\nis a refusal because the claim is not earned until Lazarus has rechecked it;\ninstall `plugins\u002Flazarus\u002Frequirements.lock` beside this checkout. The refusal\nnames the capture that failed.\n\nEmit a deterministic statement only after verification:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" statement release \\\n  --output release-statement.json\n```\n\nThe output must be outside and must not alias the release. The command emits a\ncanonical unsigned in-toto Statement v1 with one logical release subject, one\nsubject per manifest component, exact component metadata, and every capture's\ndeclared scope, coverage status and counts, unsupported collections and gaps.\nIt includes one passed Alexandria offline-verification claim bound to the\nrelease digest and an empty command list.\nA single statement must stay within Ariadne's 8,388,608-byte input limit and\nits 262,144-character key budget. A release past either bound is refused before\nthe output path is prepared, by a line naming the bound and a line pointing to\n`--parts`, so a successful output stays inside Ariadne's default reader bounds.\n\nWrite such a release as a part set instead:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" statement release \\\n  --parts release-statement-parts\n```\n\nThe target directory must be absent. A directory beside it named\n`.\u003Cname>.tmp-\u003C16 hexadecimal characters>`, with the target's own name and\nexactly 16 lowercase hexadecimal characters, refuses the run by name until it\nis removed by hand, whatever made it. The target receives `index.json` and\n`part\u002Fpart-\u003Ck>.json`, each within 6,225,920 bytes and 262,144 key characters,\nso Ariadne reads each file with its default bounds, bare or inside a DSSE\nenvelope whose own fields and signatures fit 87,380 bytes. Each part carries\na contiguous run of components and their captures; the index binds every part\nby SHA-256. `--parts` refuses a release within both\nsingle bounds and names `--output`, so each release has one statement form.\nA holder checks the set with Ariadne `verify` on the index and on each part,\nand each part's SHA-256 against its index subject. The\n[release statement guide](..\u002F..\u002Fdocs\u002Frelease-statements.md#part-sets) gives\nthe wire shapes and limits.\n\nNeither form is a DSSE envelope, and Alexandria does not run cosign. It\ndoes not authenticate a publisher or prove provider completeness, consensus\nfinality or canonical-chain membership. Ariadne can inspect each statement and\nrun its core gates, but these predicates are unregistered, signatures remain\nunchecked, and gates 2 and 5 remain unchecked.\n\nDerive the narrow Tabularium view into a new release:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" derive raw-release \\\n  --output derived-release\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" verify derived-release\n```\n\nDerivation first verifies the input and never changes it. Registered\nAave v4 and Clearpool mappings emit deterministic credit events and, where\nthe source supplies position state, observations. Every row names the raw\nrelease, component digest, source and context selectors, mapping rule, adapter\nversion and evidence class. Verification resolves those selectors, reruns the\nmappings and reconciles row, family, subject and coverage counts. Capture\nrenames do not change row IDs, and repayment legs remain neutral source\namounts unless the native record supplies a principal and interest split.\nDerivation stops above 100,000 rows or 64 MiB for either JSONL file.\n\nRebuild the disposable address index and query it:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" index derived-release \\\n  --output alexandria.sqlite\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" query \\\n  --index alexandria.sqlite --address 0x...\n```\n\nIndexing verifies every derived release, refuses to write inside one and\nretains the derived release, raw release, component, capture and row\nidentities. Querying opens SQLite read-only, checks its exact schema and\nlogical digest, then matches every indexed partition to its release. Equivalent\nrows shared by cumulative releases appear once. Conflicting content under one\nrow ID is refused.\n\nA zero-row result is empty only when complete capture scope covers every\nrequested address, venue, chain and time boundary and the mapping has no\nunsupported source records.\n\nProbitas can consume that evidence only when the operator passes its explicit\n`--alexandria-index` option. The translation retains the original venue and\nevidence class, combines per-chain coverage conservatively and keeps registry\nvenues absent from the archive visible as gaps. It does not infer people,\ndefaults, full repayment or a current balance.\n\nTo exercise the whole path without a network, run the fixed demonstration from\nthe repository root:\n\n```bash\noutput=\"$(mktemp -d)\u002Fcredit-history-v0\"\npython3 plugins\u002Falexandria\u002Fexamples\u002Fcredit-history-v0\u002Fdemo.py build --output \"$output\"\npython3 plugins\u002Falexandria\u002Fexamples\u002Fcredit-history-v0\u002Fdemo.py verify \"$output\"\n```\n\nThe plan pins existing Aave v4 and Clearpool files by digest. The result is a\nreproducibility fixture, not a production corpus. Its Aave v4 source remains\nprovider-reported and its Clearpool source remains subject-scoped with unknown\nfinality.\n\n## Compound v3 Phase 0\n\nThe [Compound method-proof example](..\u002F..\u002Fexamples\u002Fcompound-v3-phase0-v0\u002FREADME.md)\npins all 28 production Comet deployments at one upstream commit and preserves\ntwo Ethereum USDC transactions. Its checker binds old-state access, nested\ncalls, the proxy implementation and code, transaction-start storage, ordered\nstorage writes and a provider-reported finalized boundary.\n\nGenerate the fixed registry from a local pinned checkout, or build and check a\ncaptured source tree offline:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fcompound_v3_phase0.py\" registry \\\n  --comet-repository \u003Ccomet-checkout> --output registry.json\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fcompound_v3_phase0.py\" build \\\n  --input \u003Ccaptured-input> --output \u003Crelease>\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fcompound_v3_phase0.py\" check \u003Crelease>\n```\n\nThe separate `capture` subcommand is networked. It requires\n`ALEXANDRIA_COMPOUND_RPC_URL`, omits the endpoint and headers from the release,\nand collects only the fixed corpus. Do not describe the result as an interval\nhistory, chain proof or independent finality check.\n\n## Collect a declared lending interval\n\nFour commands, of which the first two reach a network:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fusdc_interval.py\" collect --plan plan.json --staging staging\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fusdc_interval.py\" reconcile --plan plan.json --staging staging \\\n  --provider-class \"\u003Cnon-secret class>\"\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fusdc_interval.py\" build --plan plan.json --staging staging \\\n  --registry registry.json --created-at \u003Ctimestamp> --output release\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fusdc_interval.py\" check release\n```\n\n`collect` dispatches on the plan's registered venue: Compound v3, Wildcat V1,\nWildcat V2 or Aave V3. Compound keeps its single proxy plan; a v2 plan names the exact\nsubject set its pinned venue registry admits. It walks bounded shards over\nthe evidence classes the plan declares, binding its end boundary\nunder the plan's named finality policy before it asks for anything. It reads its\nendpoint from `ALEXANDRIA_COMPOUND_RPC_URL` alone and writes it nowhere, and\nsends `Content-Type` and a constant `User-Agent` built from the package version.\nAn explicitly configured bearer credential may be supplied through the supported\nHTTPS transport; it never enters the plan, journal or diagnostic text. A checkpoint is\nwritten only after a shard's bytes are fsynced, so a killed run resumes to\nbyte-identical journals; when a remembered boundary hash has changed, the run\nrewinds to the deepest one that still matches, and refuses a reorg deeper than\nits bounded trail. After the last shard it makes its opening reads -- the\ninterval's first block, the implementation slot at that block and at each\nupgrade block, and each implementation's runtime code -- into a fourth\n`epoch-evidence` journal under the same checkpoint rule, so a `finalized` or\n`safe` plan now earns that finality class on every evidence scope with both\nboundary hashes bound. Structured failure receipts name the code, shard, unresolved range and provider\nclass without copying provider or transport text. CLI error strings may instead\nname only a read or shard; they do not universally include the provider class.\n\n`reconcile` runs the finished interval past a second provider and records\nagreement or dispute over boundary hashes, ordered transaction hashes and every\nlog's identity tuple. That tuple does not include the log's transaction index.\nIt settles nothing: a disputed shard becomes `partial` or `failed`, both\nproviders' bytes are kept, and a provider that cannot answer leaves the\ninterval `unreconciled`.\n\nBoth network commands take three concurrency limits. `--concurrency` (1 to 8,\ndefault 4) sets how many shards fetch at once; as many again may wait fetched\nbehind a slower lower shard, and commits stay in shard order.\n`--trace-concurrency` (1 to 16, default 4) is one total across every shard in\nflight, not a window per shard. `--rpc-concurrency` (1 to 8, default 8) caps\nevery request, traces included, so a trace limit above it adds no calls. Past\n512 MiB of request and response bytes, `collect` starts no new shard; shards\nalready started finish and commit, up to a hard stop at 1 GiB.\n\n`build` emits an ordinary capture plan and calls `ingest`; `check` verifies the\nrelease offline, re-derives every shard's record counts from the journals, and\nre-hashes each implementation's runtime code out of the `implementation-code`\ncomponent. Implementation epochs come from `discover_epochs` over every\npreserved proxy log, the EIP-1967 slot reads and the runtime code reads in the\n`epoch-evidence` journal, and are bound by code hash because the pinned\n`CometExt.version()` returns the constant string `0`. `build` takes no `--epochs`\ntable: there is nothing left for an operator to supply.\n\nA new release carries `alexandria-interval-receipt\u002Fv2`. Its epochs tile\nexclusive positions of block, transaction index and log index,\n`log_attributions` names each log's owner, and `check` re-derives both. An\nordinary log inside the upgrade transaction, an upgrade in the interval's first\nblock and two upgrades in one block refuse. A v1 release keeps its block-only\nmeaning, and an owner is an inference from the announcement's position, not\nproof of the emitting implementation.\n\nWildcat releases carry one epoch table per subject, checked against that\nvenue's registry and preserved code. A subject with no sourced deployment block\nuses preserved first-code evidence and retains the deployment-block gap.\nEach subject has its own epoch ceiling; the estate's sum is not the limit.\nBoth mainnet captures use traces selected by matching subject logs, so a\ntransaction with no such log is outside their trace coverage. Bounded concurrent\nrequests share one global cap; completed shards alone advance the checkpoint.\nThe [combined demonstration](..\u002F..\u002Fexamples\u002Fwildcat-estates-interval-v0\u002FREADME.md)\nrebuilds both estates and Compound without opening a Python socket.\n\nA `wildcat-v2` plan may declare wrapper instances in `wildcat_v2_wrappers`,\n`{\"rule\": \"wrapper-deployed\u002Fv1\", \"subjects\": [...]}`, when its `subjects`\ninclude the registry's wrapper factory. The declared wrappers join every\nshard's address filter. After the shards, each one needs a preserved\n`WrapperDeployed` log from that factory that names it and a registry market.\nIts one epoch opens at that log's block, where the block's header and the\nwrapper's runtime code are read into `epoch-evidence`. The release binds the\nderived set as the `wrapper-subjects` component. `check` re-derives it byte\nfor byte and lists each wrapper, a declared one with its epoch and attributed\nlog count. Build and `check` refuse by name an unknown market, a declared\nwrapper with no `WrapperDeployed`, a declared wrapper's `WrapperDeployed` log\nwhose block hash differs from the header read of its block, a declared\nwrapper with no code at its first block and a wrapper log before its epoch. A\ndiscovered wrapper the plan omits stays listed, and none of its logs is\nrequested. Wrapper immutability and single-factory provenance are not\nestablished, and every wrapper release says so. Only constructed data has\nexercised this path.\n\nThe `aave-v3` venue captures the Ethereum main market's 356 subjects as twelve\nsegment releases, each a plan pinned by digest in reviewed code. Its epochs are\nchosen per subject by registry role, and only this venue owns an ordinary log in\nits upgrade transaction by position. The twelve staging trees are held outside\nthe repository as one archive bound by one committed manifest;\n[`aave-v3-interval-v0`](..\u002F..\u002Fexamples\u002Faave-v3-interval-v0\u002FREADME.md)\n`verify-preserved` checks that metadata and rebuilds nothing, and the\n[proof](..\u002F..\u002Fdocs\u002Faave-v3-interval\u002Fproof.md) records each segment's results.\n\nA subject-set plan that declares `shards_per_component` may also declare\n`\"log_attribution_parts\": \"journal-ranges\"`. Its attribution rows then leave\nthe epoch table: part `k` is the component `log-attributions.\u003Ck>` and holds the\nrows of every preserved log in the plan's `k`th journal range, the range\n`logs.\u003Ck>` covers. Such a release carries `alexandria-interval-receipt\u002Fv4`,\nwhich lists each part with its shards and row count, and `check` compares every\npart with the rows it re-derives for that range. `build` and `check` refuse a\npart above 67,108,864 bytes or 2,000,000 nodes. A release holds at most 16,384\ncomponents and 16,384 captures. A plan without the field builds the same bytes\nit built before. The\n[collector document](..\u002F..\u002Fdocs\u002Fusdc-interval-collector.md#splitting-the-log-attributions-into-parts)\nstates the rule in full.\n\nRead [the collector document](..\u002F..\u002Fdocs\u002Fusdc-interval-collector.md) for the\nfinality, epoch and reconciliation boundaries, run\n[`examples\u002Fusdc-interval-v0`](..\u002F..\u002Fexamples\u002Fusdc-interval-v0\u002FREADME.md) to see\nthe whole path offline over synthetic fixtures,\n[`examples\u002Fusdc-interval-epochs-v0`](..\u002F..\u002Fexamples\u002Fusdc-interval-epochs-v0\u002FREADME.md)\nto see positional owners over a synthetic upgrade block, and\n[`examples\u002Fusdc-interval-live-v0`](..\u002F..\u002Fexamples\u002Fusdc-interval-live-v0\u002FREADME.md)\nto rebuild the preserved live Ethereum mainnet interval. This release\nestablishes no publisher identity, no provider completeness, no consensus\nfinality and no canonical-chain membership, and derives no credit event;\nTabularium owns the separate canonical mapping. The\n[harvest specification](..\u002F..\u002Fdocs\u002Fcompound-v3-harvest.md) still describes more\nthan this collector covers: one market of the pin's 28, and no state call or\nconfiguration read.\n\nRead the [study](..\u002F..\u002Fdocs\u002Fstudy.md) for the selected construction and the\n[runbook](..\u002F..\u002Fdocs\u002Frunbook.md) for the implementation boundaries.\n\n## Settled boundary\n\n1. Raw objects retain their exact bytes and native shapes.\n2. Release manifests bind components to capture scope and coverage.\n3. Tabularium mappings own the meaning of derived credit events and position\n   observations.\n4. SQLite is a disposable address index, not release evidence.\n5. Probitas retains the original venue on every archive-backed record and\n   keeps unharvested venue coverage visible.\n6. Lazarus fixtures may support selected observations but are not a universal\n   archive payload.\n\nA digest can establish that bytes match a manifest. It does not establish who\npublished them, that an indexer captured a complete chain history, or that a\nreported block is canonical. Those claims require separate evidence.\n\n## Promise Machine contract\n\n### alexandria-raw-release\n\n- Promise: A successful `ingest` followed by `verify` preserves the named local source bytes and binds them to the canonical manifest, release identity, declared capture boundary, coverage and gaps.\n- Evidence: The capture plan, copied objects, canonical manifest, printed release id and a passing `alexandria.py verify` result for the same release directory.\n- Evidence classes: recorded, checked, recomputed\n- Boundary: The result does not establish publisher identity, source completeness, chain finality or any fact outside the declared capture scope.\n- Authorises: Retention or hand-off of the verified raw release as a bounded preservation artefact.\n- Consequence: 2\n- Refuses: Installing, moving or describing a release as verified when a source path, digest, byte count, scope, coverage declaration or tree-membership check is absent or fails.\n- Recovery: Inspect the named verification failure, repair the plan or source set, ingest into a new output directory and rerun verification.\n- Exceptions: none\n\n### alexandria-derived-view\n\n- Promise: A successful `derive` followed by `verify` reproduces every emitted credit event and observation from a verified raw release under the named registered mapping and reconciles its selectors and counts.\n- Evidence: The verified raw release, adapter and mapping versions, derived JSONL, coverage records and a passing verification that reruns the mapping.\n- Evidence classes: recorded, checked, recomputed\n- Boundary: The view preserves venue and evidence classes; it does not infer default, full repayment, current balance, source completeness or universal event meaning.\n- Authorises: Use of the verified derived release as bounded input to Tabularium or an explicitly archive-backed Probitas collection.\n- Consequence: 2\n- Refuses: Using rows whose source selectors do not resolve, whose counts conflict, whose mapping is unknown or whose raw release did not verify.\n- Recovery: Inspect the failed selector, mapping or count, correct the adapter or source release without changing published evidence, derive a new release and verify it.\n- Exceptions: none\n\n### alexandria-release-statement\n\n- Promise: A successful `statement` emits a canonical unsigned in-toto Statement v1 that exactly projects a verified Alexandria release, every component digest and every capture's declared scope, coverage and gaps.\n- Evidence: The verified release, canonical statement bytes, exact release and component subject set, Alexandria predicate projection and successful command receipt for the same logical release digest.\n- Evidence classes: recorded, checked, recomputed\n- Boundary: The output has no DSSE envelope or signature check, the Alexandria predicate is unregistered in Ariadne, and the result does not establish publisher identity, provider completeness, consensus finality or canonical-chain membership.\n- Authorises: Hand-off of the unsigned statement bytes for Ariadne core-gate inspection or downstream signing without upgrading their evidence claims.\n- Consequence: 1\n- Refuses: Emitting from a release that does not verify, emitting bytes above Ariadne's default bounded-input limit, writing inside or through an alias of that release, omitting or changing a subject or predicate field, or describing unchecked signatures or predicate-owned gates as passed.\n- Recovery: Inspect the verification, projection or output-confinement failure, repair the release or choose a safe external output, rerun `statement` and retain only the successful replacement.\n- Exceptions: none\n\n### alexandria-address-query\n\n- Promise: A successful `index` and `query` returns only rows from verified derived partitions whose identities and logical digest match the disposable index.\n- Evidence: The verified derived releases, SQLite schema and logical digest checks, partition identities and the exact query result for the requested address.\n- Evidence classes: checked, recomputed\n- Boundary: SQLite is not release evidence, and zero rows does not mean a clean history unless complete declared coverage spans every requested address, venue, chain and time boundary with no unsupported records.\n- Authorises: Presentation or downstream use of the source-bound query result with its coverage and gaps kept visible.\n- Consequence: 1\n- Refuses: Querying a mutable or mismatched index, collapsing conflicting rows, or presenting an uncovered zero-row result as absence of activity.\n- Recovery: Rebuild the disposable index from verified releases, rerun the query and report any remaining coverage gap instead of a clean result.\n- Exceptions: none\n\n### alexandria-compound-method-proof\n\n- Promise: A successful Compound Phase 0 `check` binds the fixed registry and captured transaction witness to the named Comet revision, proxy implementation, code, old-state reads, calls and ordered writes under the implemented method.\n- Evidence: The pinned registry, captured input, built release and passing `compound_v3_phase0.py check` result, including the proof-checked state relation and separately labelled provider records.\n- Evidence classes: recorded, checked, recomputed, proved: EIP-1186 state relation\n- Boundary: The result covers the fixed corpus and implemented method only; it is not an interval history, independent canonical-chain proof, canonical Tabularium mapping or proof of receipts, logs or traces.\n- Authorises: Use of the checked Phase 0 witness as bounded method evidence for the named transactions.\n- Consequence: 1\n- Refuses: Generalising the witness to another deployment, transaction, interval, implementation, layout or evidence class.\n- Recovery: Inspect the named registry, implementation, state, call, write or selector mismatch, recapture under an amended fixed plan and rerun the check.\n- Exceptions: none\n\n### alexandria-proof-backed-state\n\n- Promise: A successful `verify` of a release whose capture claims `proof-backed-state` establishes that the capture component is an embedded Lazarus fixture manifest, every file it names is present by digest, Lazarus's offline verifier accepted the reconstructed fixture, and the capture's chain, block, finality and subjects claim no more than that fixture proves.\n- Evidence: The verified Alexandria manifest and components, digest-and-byte-count fixture reconstruction, Lazarus verification report, bound capture fields and executable P\u002FM\u002FO\u002FR\u002FS cases in `plugins\u002Falexandria\u002Ftests\u002Ftest_proof_backed.py`.\n- Evidence classes: recorded, checked, recomputed, proved: EIP-1186 state relation\n- Boundary: The result does not establish canonical-chain membership, provider independence, source completeness or anything outside the named proof targets at the bound block; it borrows no verdict because Alexandria reruns Lazarus instead of trusting one.\n- Authorises: Retaining or handing off the verified capture's `proof-backed-state` class for its named subjects at the bound block.\n- Consequence: 1\n- Refuses: Describing such a capture as verified when the sibling verifier is absent, the fixture is incomplete or altered, or the claim exceeds the proof.\n- Recovery: Repair the plan or fixture, ingest into a new output and rerun `verify`.\n- Exceptions: none\n","\u003Cp align=\"center\">\n  \u003Cimg src=\"..\u002F..\u002Fassets\u002Fcharacters\u002Falexandria.png\" width=\"1200\">\n\u003C\u002Fp>\n\n# Alexandria\n\n## Frontier\n\nAlexandria owns its own preservation and credit-view frontier, not Hexaemeron's delivery or\nSolidity frontier. Its version, held target, next job, and maturity\nstate live in [EVOLUTION.md](EVOLUTION.md). Do not recommend or run\nanother frontier pass after that ledger becomes mature.\n\n\u003C!-- marketplace-context:start -->\n## Where this sits\n\nAlexandria preserves heterogeneous lending captures byte for byte, then exposes only the source-bound credit view a reviewed mapping can defend.\n\n**Current frontier.** Ordinary builds now emit `alexandria-interval-receipt\u002Fv2`, which attributes each preserved proxy log to an implementation epoch by block, transaction index and log index, and `check` re-derives every owner offline; reconciliation still compares a log without its transaction index, so a second provider that reports a different index for the same log records `agreed`.\n\u003C!-- marketplace-context:end -->\n\nAlexandria is the archive and catalogue behind durable lending-protocol\nresearch. Raw captures remain unchanged. Tabularium owns their interpretation\nas venue-qualified credit events, and Probitas may consume those events in a\ndossier. Ariadne may bind a finished release to its evidence. Lazarus is the\nneighbouring specialist for the finite historical Ethereum state and exact RPC\ntraffic one application test needs; it does not replace Alexandria's dataset\nboundary.\n\nSynkrisis compares validated agent-run observations through a checked cohort,\nbounded findings, a fixed report, and whole-path verification. It cannot\nreinterpret an Alexandria release, treat release rows as run observations, or\nauthorise a new capture.\n\n`$SKILL_DIR` is the directory holding this file. The command lives at\n`$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py`; resolve it from where you loaded this\nskill.\n\n## Day to day\n\n**Research.** A protocol endpoint or hosted indexer may disappear. Preserve\nthe exact response, its capture scope and its gaps before the only cheap copy\nis gone.\n\n**Credit.** A counterparty record needs events, position observations and an\naccount of the venues and intervals checked. An empty result without coverage\nis not a clean history.\n\n**Data engineering.** Raw GraphQL responses, chain logs and replay fixtures\ndo not need one payload schema. Bind their bytes in one manifest, then build a\nsmall, versioned credit view that points back to them.\n\n## Raw releases\n\nPrepare a capture plan using the contracts under\n`$SKILL_DIR\u002F..\u002F..\u002Fschemas\u002F`. Every component path is relative to the plan's\ndirectory. Every component has a `public`, `restricted` or `private` access\nclass and a `permitted`, `restricted`, `prohibited` or `unknown` redistribution\nclass. Give every capture a venue, an `eip155` chain, a non-secret source\nreference, an evidence and finality class, a full-dataset or subject-scoped\nboundary and counted coverage or a stated gap. Then ingest it:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" ingest \\\n  --plan capture-plan.json --output release\n```\n\nIngest reads only the plan and its confined local source files. It copies raw\nbytes unchanged into digest-derived paths, writes through a temporary sibling\ndirectory and installs the release atomically. It refuses absolute paths,\ntraversal, symlinks and replacement of a different release. Record the\n`sha256:...` release ID printed on success.\n\nVerify before using or moving a release:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" verify release\n```\n\nVerification is offline and read-only. It checks canonical manifest bytes,\nrelease identity, object paths, byte counts, digests, component access and\nredistribution classes, capture source, scope, evidence and finality classes,\ncollection counts, declared gaps, correction links and exact release-tree\nmembership. It does not establish publisher identity, source completeness or\nchain finality.\n\nA `proof-backed-state` capture earns that class only during `verify`: its\ncomponent must be a Lazarus manifest, every named fixture file must be present\nby digest and byte count, Alexandria must reconstruct and rerun Lazarus, and\nthe capture must stay within the proved block and targets. Finality must be\n`unknown` because Lazarus proves block binding but reports no finality class.\nThe scope must be subject-scoped and remain within the proof targets because a\nfinite proof set is not a full dataset. Missing Lazarus or its pinned packages\nis a refusal because the claim is not earned until Lazarus has rechecked it;\ninstall `plugins\u002Flazarus\u002Frequirements.lock` beside this checkout. The refusal\nnames the capture that failed.\n\nEmit a deterministic statement only after verification:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" statement release \\\n  --output release-statement.json\n```\n\nThe output must be outside and must not alias the release. The command emits a\ncanonical unsigned in-toto Statement v1 with one logical release subject, one\nsubject per manifest component, exact component metadata, and every capture's\ndeclared scope, coverage status and counts, unsupported collections and gaps.\nIt includes one passed Alexandria offline-verification claim bound to the\nrelease digest and an empty command list.\nA single statement must stay within Ariadne's 8,388,608-byte input limit and\nits 262,144-character key budget. A release past either bound is refused before\nthe output path is prepared, by a line naming the bound and a line pointing to\n`--parts`, so a successful output stays inside Ariadne's default reader bounds.\n\nWrite such a release as a part set instead:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" statement release \\\n  --parts release-statement-parts\n```\n\nThe target directory must be absent. A directory beside it named\n`.\u003Cname>.tmp-\u003C16 hexadecimal characters>`, with the target's own name and\nexactly 16 lowercase hexadecimal characters, refuses the run by name until it\nis removed by hand, whatever made it. The target receives `index.json` and\n`part\u002Fpart-\u003Ck>.json`, each within 6,225,920 bytes and 262,144 key characters,\nso Ariadne reads each file with its default bounds, bare or inside a DSSE\nenvelope whose own fields and signatures fit 87,380 bytes. Each part carries\na contiguous run of components and their captures; the index binds every part\nby SHA-256. `--parts` refuses a release within both\nsingle bounds and names `--output`, so each release has one statement form.\nA holder checks the set with Ariadne `verify` on the index and on each part,\nand each part's SHA-256 against its index subject. The\n[release statement guide](..\u002F..\u002Fdocs\u002Frelease-statements.md#part-sets) gives\nthe wire shapes and limits.\n\nNeither form is a DSSE envelope, and Alexandria does not run cosign. It\ndoes not authenticate a publisher or prove provider completeness, consensus\nfinality or canonical-chain membership. Ariadne can inspect each statement and\nrun its core gates, but these predicates are unregistered, signatures remain\nunchecked, and gates 2 and 5 remain unchecked.\n\nDerive the narrow Tabularium view into a new release:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" derive raw-release \\\n  --output derived-release\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" verify derived-release\n```\n\nDerivation first verifies the input and never changes it. Registered\nAave v4 and Clearpool mappings emit deterministic credit events and, where\nthe source supplies position state, observations. Every row names the raw\nrelease, component digest, source and context selectors, mapping rule, adapter\nversion and evidence class. Verification resolves those selectors, reruns the\nmappings and reconciles row, family, subject and coverage counts. Capture\nrenames do not change row IDs, and repayment legs remain neutral source\namounts unless the native record supplies a principal and interest split.\nDerivation stops above 100,000 rows or 64 MiB for either JSONL file.\n\nRebuild the disposable address index and query it:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" index derived-release \\\n  --output alexandria.sqlite\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Falexandria.py\" query \\\n  --index alexandria.sqlite --address 0x...\n```\n\nIndexing verifies every derived release, refuses to write inside one and\nretains the derived release, raw release, component, capture and row\nidentities. Querying opens SQLite read-only, checks its exact schema and\nlogical digest, then matches every indexed partition to its release. Equivalent\nrows shared by cumulative releases appear once. Conflicting content under one\nrow ID is refused.\n\nA zero-row result is empty only when complete capture scope covers every\nrequested address, venue, chain and time boundary and the mapping has no\nunsupported source records.\n\nProbitas can consume that evidence only when the operator passes its explicit\n`--alexandria-index` option. The translation retains the original venue and\nevidence class, combines per-chain coverage conservatively and keeps registry\nvenues absent from the archive visible as gaps. It does not infer people,\ndefaults, full repayment or a current balance.\n\nTo exercise the whole path without a network, run the fixed demonstration from\nthe repository root:\n\n```bash\noutput=\"$(mktemp -d)\u002Fcredit-history-v0\"\npython3 plugins\u002Falexandria\u002Fexamples\u002Fcredit-history-v0\u002Fdemo.py build --output \"$output\"\npython3 plugins\u002Falexandria\u002Fexamples\u002Fcredit-history-v0\u002Fdemo.py verify \"$output\"\n```\n\nThe plan pins existing Aave v4 and Clearpool files by digest. The result is a\nreproducibility fixture, not a production corpus. Its Aave v4 source remains\nprovider-reported and its Clearpool source remains subject-scoped with unknown\nfinality.\n\n## Compound v3 Phase 0\n\nThe [Compound method-proof example](..\u002F..\u002Fexamples\u002Fcompound-v3-phase0-v0\u002FREADME.md)\npins all 28 production Comet deployments at one upstream commit and preserves\ntwo Ethereum USDC transactions. Its checker binds old-state access, nested\ncalls, the proxy implementation and code, transaction-start storage, ordered\nstorage writes and a provider-reported finalized boundary.\n\nGenerate the fixed registry from a local pinned checkout, or build and check a\ncaptured source tree offline:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fcompound_v3_phase0.py\" registry \\\n  --comet-repository \u003Ccomet-checkout> --output registry.json\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fcompound_v3_phase0.py\" build \\\n  --input \u003Ccaptured-input> --output \u003Crelease>\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fcompound_v3_phase0.py\" check \u003Crelease>\n```\n\nThe separate `capture` subcommand is networked. It requires\n`ALEXANDRIA_COMPOUND_RPC_URL`, omits the endpoint and headers from the release,\nand collects only the fixed corpus. Do not describe the result as an interval\nhistory, chain proof or independent finality check.\n\n## Collect a declared lending interval\n\nFour commands, of which the first two reach a network:\n\n```bash\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fusdc_interval.py\" collect --plan plan.json --staging staging\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fusdc_interval.py\" reconcile --plan plan.json --staging staging \\\n  --provider-class \"\u003Cnon-secret class>\"\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fusdc_interval.py\" build --plan plan.json --staging staging \\\n  --registry registry.json --created-at \u003Ctimestamp> --output release\npython3 \"$SKILL_DIR\u002F..\u002F..\u002Fscripts\u002Fusdc_interval.py\" check release\n```\n\n`collect` dispatches on the plan's registered venue: Compound v3, Wildcat V1,\nWildcat V2 or Aave V3. Compound keeps its single proxy plan; a v2 plan names the exact\nsubject set its pinned venue registry admits. It walks bounded shards over\nthe evidence classes the plan declares, binding its end boundary\nunder the plan's named finality policy before it asks for anything. It reads its\nendpoint from `ALEXANDRIA_COMPOUND_RPC_URL` alone and writes it nowhere, and\nsends `Content-Type` and a constant `User-Agent` built from the package version.\nAn explicitly configured bearer credential may be supplied through the supported\nHTTPS transport; it never enters the plan, journal or diagnostic text. A checkpoint is\nwritten only after a shard's bytes are fsynced, so a killed run resumes to\nbyte-identical journals; when a remembered boundary hash has changed, the run\nrewinds to the deepest one that still matches, and refuses a reorg deeper than\nits bounded trail. After the last shard it makes its opening reads -- the\ninterval's first block, the implementation slot at that block and at each\nupgrade block, and each implementation's runtime code -- into a fourth\n`epoch-evidence` journal under the same checkpoint rule, so a `finalized` or\n`safe` plan now earns that finality class on every evidence scope with both\nboundary hashes bound. Structured failure receipts name the code, shard, unresolved range and provider\nclass without copying provider or transport text. CLI error strings may instead\nname only a read or shard; they do not universally include the provider class.\n\n`reconcile` runs the finished interval past a second provider and records\nagreement or dispute over boundary hashes, ordered transaction hashes and every\nlog's identity tuple. That tuple does not include the log's transaction index.\nIt settles nothing: a disputed shard becomes `partial` or `failed`, both\nproviders' bytes are kept, and a provider that cannot answer leaves the\ninterval `unreconciled`.\n\nBoth network commands take three concurrency limits. `--concurrency` (1 to 8,\ndefault 4) sets how many shards fetch at once; as many again may wait fetched\nbehind a slower lower shard, and commits stay in shard order.\n`--trace-concurrency` (1 to 16, default 4) is one total across every shard in\nflight, not a window per shard. `--rpc-concurrency` (1 to 8, default 8) caps\nevery request, traces included, so a trace limit above it adds no calls. Past\n512 MiB of request and response bytes, `collect` starts no new shard; shards\nalready started finish and commit, up to a hard stop at 1 GiB.\n\n`build` emits an ordinary capture plan and calls `ingest`; `check` verifies the\nrelease offline, re-derives every shard's record counts from the journals, and\nre-hashes each implementation's runtime code out of the `implementation-code`\ncomponent. Implementation epochs come from `discover_epochs` over every\npreserved proxy log, the EIP-1967 slot reads and the runtime code reads in the\n`epoch-evidence` journal, and are bound by code hash because the pinned\n`CometExt.version()` returns the constant string `0`. `build` takes no `--epochs`\ntable: there is nothing left for an operator to supply.\n\nA new release carries `alexandria-interval-receipt\u002Fv2`. Its epochs tile\nexclusive positions of block, transaction index and log index,\n`log_attributions` names each log's owner, and `check` re-derives both. An\nordinary log inside the upgrade transaction, an upgrade in the interval's first\nblock and two upgrades in one block refuse. A v1 release keeps its block-only\nmeaning, and an owner is an inference from the announcement's position, not\nproof of the emitting implementation.\n\nWildcat releases carry one epoch table per subject, checked against that\nvenue's registry and preserved code. A subject with no sourced deployment block\nuses preserved first-code evidence and retains the deployment-block gap.\nEach subject has its own epoch ceiling; the estate's sum is not the limit.\nBoth mainnet captures use traces selected by matching subject logs, so a\ntransaction with no such log is outside their trace coverage. Bounded concurrent\nrequests share one global cap; completed shards alone advance the checkpoint.\nThe [combined demonstration](..\u002F..\u002Fexamples\u002Fwildcat-estates-interval-v0\u002FREADME.md)\nrebuilds both estates and Compound without opening a Python socket.\n\nA `wildcat-v2` plan may declare wrapper instances in `wildcat_v2_wrappers`,\n`{\"rule\": \"wrapper-deployed\u002Fv1\", \"subjects\": [...]}`, when its `subjects`\ninclude the registry's wrapper factory. The declared wrappers join every\nshard's address filter. After the shards, each one needs a preserved\n`WrapperDeployed` log from that factory that names it and a registry market.\nIts one epoch opens at that log's block, where the block's header and the\nwrapper's runtime code are read into `epoch-evidence`. The release binds the\nderived set as the `wrapper-subjects` component. `check` re-derives it byte\nfor byte and lists each wrapper, a declared one with its epoch and attributed\nlog count. Build and `check` refuse by name an unknown market, a declared\nwrapper with no `WrapperDeployed`, a declared wrapper's `WrapperDeployed` log\nwhose block hash differs from the header read of its block, a declared\nwrapper with no code at its first block and a wrapper log before its epoch. A\ndiscovered wrapper the plan omits stays listed, and none of its logs is\nrequested. Wrapper immutability and single-factory provenance are not\nestablished, and every wrapper release says so. Only constructed data has\nexercised this path.\n\nThe `aave-v3` venue captures the Ethereum main market's 356 subjects as twelve\nsegment releases, each a plan pinned by digest in reviewed code. Its epochs are\nchosen per subject by registry role, and only this venue owns an ordinary log in\nits upgrade transaction by position. The twelve staging trees are held outside\nthe repository as one archive bound by one committed manifest;\n[`aave-v3-interval-v0`](..\u002F..\u002Fexamples\u002Faave-v3-interval-v0\u002FREADME.md)\n`verify-preserved` checks that metadata and rebuilds nothing, and the\n[proof](..\u002F..\u002Fdocs\u002Faave-v3-interval\u002Fproof.md) records each segment's results.\n\nA subject-set plan that declares `shards_per_component` may also declare\n`\"log_attribution_parts\": \"journal-ranges\"`. Its attribution rows then leave\nthe epoch table: part `k` is the component `log-attributions.\u003Ck>` and holds the\nrows of every preserved log in the plan's `k`th journal range, the range\n`logs.\u003Ck>` covers. Such a release carries `alexandria-interval-receipt\u002Fv4`,\nwhich lists each part with its shards and row count, and `check` compares every\npart with the rows it re-derives for that range. `build` and `check` refuse a\npart above 67,108,864 bytes or 2,000,000 nodes. A release holds at most 16,384\ncomponents and 16,384 captures. A plan without the field builds the same bytes\nit built before. The\n[collector document](..\u002F..\u002Fdocs\u002Fusdc-interval-collector.md#splitting-the-log-attributions-into-parts)\nstates the rule in full.\n\nRead [the collector document](..\u002F..\u002Fdocs\u002Fusdc-interval-collector.md) for the\nfinality, epoch and reconciliation boundaries, run\n[`examples\u002Fusdc-interval-v0`](..\u002F..\u002Fexamples\u002Fusdc-interval-v0\u002FREADME.md) to see\nthe whole path offline over synthetic fixtures,\n[`examples\u002Fusdc-interval-epochs-v0`](..\u002F..\u002Fexamples\u002Fusdc-interval-epochs-v0\u002FREADME.md)\nto see positional owners over a synthetic upgrade block, and\n[`examples\u002Fusdc-interval-live-v0`](..\u002F..\u002Fexamples\u002Fusdc-interval-live-v0\u002FREADME.md)\nto rebuild the preserved live Ethereum mainnet interval. This release\nestablishes no publisher identity, no provider completeness, no consensus\nfinality and no canonical-chain membership, and derives no credit event;\nTabularium owns the separate canonical mapping. The\n[harvest specification](..\u002F..\u002Fdocs\u002Fcompound-v3-harvest.md) still describes more\nthan this collector covers: one market of the pin's 28, and no state call or\nconfiguration read.\n\nRead the [study](..\u002F..\u002Fdocs\u002Fstudy.md) for the selected construction and the\n[runbook](..\u002F..\u002Fdocs\u002Frunbook.md) for the implementation boundaries.\n\n## Settled boundary\n\n1. Raw objects retain their exact bytes and native shapes.\n2. Release manifests bind components to capture scope and coverage.\n3. Tabularium mappings own the meaning of derived credit events and position\n   observations.\n4. SQLite is a disposable address index, not release evidence.\n5. Probitas retains the original venue on every archive-backed record and\n   keeps unharvested venue coverage visible.\n6. Lazarus fixtures may support selected observations but are not a universal\n   archive payload.\n\nA digest can establish that bytes match a manifest. It does not establish who\npublished them, that an indexer captured a complete chain history, or that a\nreported block is canonical. Those claims require separate evidence.\n\n## Promise Machine contract\n\n### alexandria-raw-release\n\n- Promise: A successful `ingest` followed by `verify` preserves the named local source bytes and binds them to the canonical manifest, release identity, declared capture boundary, coverage and gaps.\n- Evidence: The capture plan, copied objects, canonical manifest, printed release id and a passing `alexandria.py verify` result for the same release directory.\n- Evidence classes: recorded, checked, recomputed\n- Boundary: The result does not establish publisher identity, source completeness, chain finality or any fact outside the declared capture scope.\n- Authorises: Retention or hand-off of the verified raw release as a bounded preservation artefact.\n- Consequence: 2\n- Refuses: Installing, moving or describing a release as verified when a source path, digest, byte count, scope, coverage declaration or tree-membership check is absent or fails.\n- Recovery: Inspect the named verification failure, repair the plan or source set, ingest into a new output directory and rerun verification.\n- Exceptions: none\n\n### alexandria-derived-view\n\n- Promise: A successful `derive` followed by `verify` reproduces every emitted credit event and observation from a verified raw release under the named registered mapping and reconciles its selectors and counts.\n- Evidence: The verified raw release, adapter and mapping versions, derived JSONL, coverage records and a passing verification that reruns the mapping.\n- Evidence classes: recorded, checked, recomputed\n- Boundary: The view preserves venue and evidence classes; it does not infer default, full repayment, current balance, source completeness or universal event meaning.\n- Authorises: Use of the verified derived release as bounded input to Tabularium or an explicitly archive-backed Probitas collection.\n- Consequence: 2\n- Refuses: Using rows whose source selectors do not resolve, whose counts conflict, whose mapping is unknown or whose raw release did not verify.\n- Recovery: Inspect the failed selector, mapping or count, correct the adapter or source release without changing published evidence, derive a new release and verify it.\n- Exceptions: none\n\n### alexandria-release-statement\n\n- Promise: A successful `statement` emits a canonical unsigned in-toto Statement v1 that exactly projects a verified Alexandria release, every component digest and every capture's declared scope, coverage and gaps.\n- Evidence: The verified release, canonical statement bytes, exact release and component subject set, Alexandria predicate projection and successful command receipt for the same logical release digest.\n- Evidence classes: recorded, checked, recomputed\n- Boundary: The output has no DSSE envelope or signature check, the Alexandria predicate is unregistered in Ariadne, and the result does not establish publisher identity, provider completeness, consensus finality or canonical-chain membership.\n- Authorises: Hand-off of the unsigned statement bytes for Ariadne core-gate inspection or downstream signing without upgrading their evidence claims.\n- Consequence: 1\n- Refuses: Emitting from a release that does not verify, emitting bytes above Ariadne's default bounded-input limit, writing inside or through an alias of that release, omitting or changing a subject or predicate field, or describing unchecked signatures or predicate-owned gates as passed.\n- Recovery: Inspect the verification, projection or output-confinement failure, repair the release or choose a safe external output, rerun `statement` and retain only the successful replacement.\n- Exceptions: none\n\n### alexandria-address-query\n\n- Promise: A successful `index` and `query` returns only rows from verified derived partitions whose identities and logical digest match the disposable index.\n- Evidence: The verified derived releases, SQLite schema and logical digest checks, partition identities and the exact query result for the requested address.\n- Evidence classes: checked, recomputed\n- Boundary: SQLite is not release evidence, and zero rows does not mean a clean history unless complete declared coverage spans every requested address, venue, chain and time boundary with no unsupported records.\n- Authorises: Presentation or downstream use of the source-bound query result with its coverage and gaps kept visible.\n- Consequence: 1\n- Refuses: Querying a mutable or mismatched index, collapsing conflicting rows, or presenting an uncovered zero-row result as absence of activity.\n- Recovery: Rebuild the disposable index from verified releases, rerun the query and report any remaining coverage gap instead of a clean result.\n- Exceptions: none\n\n### alexandria-compound-method-proof\n\n- Promise: A successful Compound Phase 0 `check` binds the fixed registry and captured transaction witness to the named Comet revision, proxy implementation, code, old-state reads, calls and ordered writes under the implemented method.\n- Evidence: The pinned registry, captured input, built release and passing `compound_v3_phase0.py check` result, including the proof-checked state relation and separately labelled provider records.\n- Evidence classes: recorded, checked, recomputed, proved: EIP-1186 state relation\n- Boundary: The result covers the fixed corpus and implemented method only; it is not an interval history, independent canonical-chain proof, canonical Tabularium mapping or proof of receipts, logs or traces.\n- Authorises: Use of the checked Phase 0 witness as bounded method evidence for the named transactions.\n- Consequence: 1\n- Refuses: Generalising the witness to another deployment, transaction, interval, implementation, layout or evidence class.\n- Recovery: Inspect the named registry, implementation, state, call, write or selector mismatch, recapture under an amended fixed plan and rerun the check.\n- Exceptions: none\n\n### alexandria-proof-backed-state\n\n- Promise: A successful `verify` of a release whose capture claims `proof-backed-state` establishes that the capture component is an embedded Lazarus fixture manifest, every file it names is present by digest, Lazarus's offline verifier accepted the reconstructed fixture, and the capture's chain, block, finality and subjects claim no more than that fixture proves.\n- Evidence: The verified Alexandria manifest and components, digest-and-byte-count fixture reconstruction, Lazarus verification report, bound capture fields and executable P\u002FM\u002FO\u002FR\u002FS cases in `plugins\u002Falexandria\u002Ftests\u002Ftest_proof_backed.py`.\n- Evidence classes: recorded, checked, recomputed, proved: EIP-1186 state relation\n- Boundary: The result does not establish canonical-chain membership, provider independence, source completeness or anything outside the named proof targets at the bound block; it borrows no verdict because Alexandria reruns Lazarus instead of trusting one.\n- Authorises: Retaining or handing off the verified capture's `proof-backed-state` class for its named subjects at the bound block.\n- Consequence: 1\n- Refuses: Describing such a capture as verified when the sibling verifier is absent, the fixture is incomplete or altered, or the claim exceeds the proof.\n- Recovery: Repair the plan or fixture, ingest into a new output and rerun `verify`.\n- Exceptions: none\n",{"name":5,"metadata":24,"description":26},{"version":25},"3.13.0","Preserve heterogeneous lending-protocol captures by digest and expose a narrow, source-bound credit view for Tabularium and Probitas. Use when the user names Alexandria or asks to archive lending data for reproducible, address-scoped credit research. Raw release and registered Aave v4 and Clearpool derivation, disposable indexing, address queries and a checked-in offline demonstration, unsigned in-toto release statements, a bounded Compound v3 Phase 0 method proof and a resumable Ethereum USDC interval collector with per-subject implementation epochs and preserved Compound, Wildcat V1, Wildcat V2 and Aave V3 mainnet intervals are available.\n","main","6b6406ecd922c4127bc688e0b5122bbe395be0eb","plugins\u002Falexandria\u002Fskills\u002Falexandria",6,"https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills","https:\u002F\u002Fgithub.com\u002Fwildcat-finance","https:\u002F\u002Fwildcat.finance","Banking, But Worse","2026-11-07 17:14:07",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":48},[38,41,44],{"tool":39,"command":40},"npx","npx skills add wildcat-finance\u002Fskills --skill alexandria",{"tool":42,"command":43},"bunx","bunx skills add wildcat-finance\u002Fskills --skill alexandria",{"tool":45,"command":46},"pnpm","pnpm dlx skills add wildcat-finance\u002Fskills --skill alexandria","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Falexandria\u002Fskills\u002Falexandria\u002FSKILL.md",true,[50,56,60],{"path":51,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":54,"sourceHtmlUrl":55},"agents\u002Fopenai.yaml","REFERENCE","参考文档",647,"https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Falexandria\u002Fskills\u002Falexandria\u002Fagents\u002Fopenai.yaml",{"path":57,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":58,"sourceHtmlUrl":59},"DEMONSTRATION.md",13360,"https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Falexandria\u002Fskills\u002Falexandria\u002FDEMONSTRATION.md",{"path":61,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":62,"sourceHtmlUrl":63},"EVOLUTION.md",17249,"https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Falexandria\u002Fskills\u002Falexandria\u002FEVOLUTION.md",[65,72,78,85,91,97],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M7APAWXHFSKVQEJBAHKASH25","example","A deliberately unsupported evidence class fixture.","https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fpromise-machine\u002Funsupported-evidence-class\u002Fplugins\u002Fexample\u002Fskills\u002Fexample\u002FSKILL.md",[],[],{"skillNo":73,"name":67,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M7APATBK93107VF5NMPA3RJH","A router fixture target.","https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fpromise-machine\u002Funresolved-router\u002Fplugins\u002Fexample\u002Fskills\u002Fexample\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M7APARV9C0MQB8WTB0GGAAMG","promise-machine","A deliberately unresolved router.","https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fpromise-machine\u002Funresolved-router\u002F.agents\u002Fskills\u002Fpromise-machine\u002FSKILL.md",[],[],{"skillNo":86,"name":67,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7APAQMDG4977HXGSKMB29KV","A deliberately unclassified skill fixture.","https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fpromise-machine\u002Funclassified-skill\u002Fplugins\u002Fexample\u002Fskills\u002Fexample\u002FSKILL.md",[],[],{"skillNo":92,"name":67,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M7APAPE6DE05A24J4BYPK6WY","A deliberately unattributed exception fixture.","https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fpromise-machine\u002Funattributed-exception\u002Fplugins\u002Fexample\u002Fskills\u002Fexample\u002FSKILL.md",[],[],{"skillNo":98,"name":67,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":19},"01M7APAKYXRTABBSN7Q637CC6M","A deliberately conflated version fixture.","https:\u002F\u002Fgithub.com\u002Fwildcat-finance\u002Fskills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fpromise-machine\u002Fpackage-as-skill-version\u002Fplugins\u002Fexample\u002Fskills\u002Fexample\u002FSKILL.md",[],[]]