Evidence report and limitations

EN source master · P08-L08 · 45 minutes estimated · needs_review

A report is useful only if a reviewer can recover the object, evidence and reasoning behind each claim. A confident headline cannot compensate for mixed contexts, missing balances or untested issuer attribution. This final P08 lesson turns narrow findings into a bounded diagnostic report.

The exercise is an asset report, not the future Wallet Intelligence program. It assesses provenance, reconciliation, uncertainty and communication. No certificate, investment instruction or runtime Passport is issued.

Begin with the question and cutoff

Define the object tuple and the question. “Review SIM-SOL-A / MINT-R at supplied S0 for identity, controls, distribution and liquidity evidence” is bounded. “Tell me whether it is safe” lacks an operational definition and invites unsupported guarantees.

Specify an as-of context separately from report generation. A report created later can describe earlier evidence. Its creation time is not the time the token state was measured. Record observation/context time, source retrieval time and report creation time as separate fields where available.

This exercise uses synthetic T+ capture/generation labels and S0 state labels. They are not UTC timestamps and must not be formatted as collected real observations. A real report would need actual available block/slot/time, finality and acquisition provenance. If wall-clock time is unavailable, preserve the returned context without guessing.

Create an evidence register

Give each record a stable locator, source, exact object, context, included fields and limitations. A claim should refer to one or more records rather than to “the dashboard.” This lets reviewers assess whether the evidence actually answers the question.

The supplied raw supply and decimals can be normalized using the same method as P08-L02; Solana getTokenSupply documents those response fields [S08-SUPPLY]. The resulting arithmetic is a calculation, not another independent source. Do not count derived metrics as corroborating observations.

A metadata update mechanism or LP position method is documented separately [S08-METADATA; S08-POOLS]. Documentation can explain a control or claim representation; it cannot authenticate invented dataset values. Bind the methodological source and the measurement record distinctly.

Classify claims before drafting the conclusion

OBSERVED means a directly inspectable observation within scope. Here it means a supplied fictional record, never a live-chain measurement. INFERRED means reasoning from records and declared assumptions. UNKNOWN marks unavailable facts. INSUFFICIENT EVIDENCE answers a tested claim the available records cannot support.

These states are not an ascending safety scale. Unknown issuer control can coexist with observed mint identity. Insufficient evidence for all-liquidity-locked can coexist with observed reserves. Do not average states into a reassuring score.

A contradiction needs its own handling. R4 reports an older assertion that freeze authority was revoked; R2 records active freeze authority F at supplied S0. For the same correctly identified standard Token Program mint and the same freeze-authority role, a valid revocation to None permanently removes that role [S08-SET]; the mint can no longer use that authority to freeze token accounts [S08-FREEZE]. These assertions must not be reconciled as ordinary later reactivation.

Use matched R2 to describe what the supplied S0 record reports, while preserving R4 and classifying the cause of the conflict as UNKNOWN / UNRESOLVED. Possible explanations include an inaccurate older assertion, a wrong mint/object or program/context, a decoding/reporting error, or a record-provenance problem. No explanation is established here. R2 does not prove why R4 conflicts, and neither record becomes independently authenticated history. The documentation establishes the mechanism, not the truth of the fixture values.

This rule is scoped to the identified mint, relevant role and standard documented authority semantics. It does not establish the absence of other controls: Token-2022 extensions, auxiliary programs or different mechanisms require their own evidence and analysis.

Reconcile coverage and denominators

State the supplied total and export coverage. A partial distribution can support observed balances and a coverage fraction, not an exhaustive owner ranking. If a reported owner balance is 2000 of 10000, its supplied share is 20%. Without complete enumeration that may be a lower bound on that owner authority's total, depending on whether more accounts exist.

A missing5000 tokens in a half-covered export is not a new holder named Other. It is unreturned inventory of unknown composition. It must not be labelled team, burn, pool or retail without evidence.

Pool reserves and withdrawal rights remain separate. A reserve row can establish a supplied pool quantity at a context. It cannot establish all claims locked when the claim ledger and locker configuration are missing. Carry that limitation into the final paragraph rather than hiding it in an appendix.

Manage freshness and inconsistent records

A retrieval-age calculation needs real times in comparable units. T+ labels in this simulation only support relative ordering, not a live freshness threshold. A report captured at T+10 and generated at T+20 contains a ten-step interval; it does not establish ten minutes or seconds.

Matching context is necessary but not sufficient. Decoders, commitment, filters and source errors still matter. If the dataset omits a fact, do not fabricate a successful provider check. If two sources depend on one upstream provider, their independence remains uncertain unless established.

Fail closed on unsupported output fields. If beneficial ownership is unknown, leave it unknown. If an update controller is known only by an account label, do not name a human. If a website declaration lacks authentication, retain declared/unverified status.

Structure the report for review

A practical report contains scope, evidence register, narrow findings, limitations and next checks. Each finding states data status, record pointers and context. Include calculations with numerator and denominator so a reviewer can reproduce them.

Do not conclude with BUY, SELL, HOLD, a profit promise or an unsupported safety verdict. A diagnostic snapshot and a prediction have different purposes. A future Diagnostic Passport may preserve observations if its runtime capabilities are verified; this Markdown exercise is not such a generated artifact.

Keep recommendations about investigation concrete: fetch full owner/account coverage, inspect LP claims and investigate conflicting authority-record provenance. These are evidence requests, not instructions to trade. The owner can review the learning content without accessing paid tools.

Preserve version and review state

Content version and lastReviewedAt describe editorial provenance. needs_review is truthful until owner/content/source review is completed. It is not a claim the token itself is unreviewed or malicious. Do not populate lastReviewedAt with the writing date and imply independent review.

A later state change may invalidate a diagnostic claim without invalidating the lesson's mathematics. Preserve the original context and append a new report version rather than rewriting a past observation to match the present. This is why scope and immutable evidence pointers matter.

• Evidence register: indexed source records supporting claims.
• As-of cutoff: state boundary addressed by the report.
• Capture time: when source evidence was retrieved.
• Coverage fraction: observed inventory divided by stated total.
• Lower bound: quantity that may omit additional holdings.
• Diagnostic snapshot: bounded observations, not a prediction.
• Review state: editorial status distinct from token classification.

ILLUSTRATIVE — FIX-P08-08. SIM-SOL-A / MINT-R. Source F-A is a fictional frozen provider export captured at synthetic T+10; report generation is T+20. State cutoff S0. No real seconds, block, slot or UTC observation was collected.

R6 is complete only for the two returned records, not for the mint. Quote precision is already normalized in R3. No other A accounts, beneficiaries, LP claims or authority-transition records are supplied.

OBSERVED in fixture: S0 identity/fields and partial balances are supplied; F-A is shared provenance for R1/R2/R3/R6.

INFERRED: Total10,000 tokens, A returned 2000, pool3000 and export coverage50%. A's returned balance is 20% of total; its exhaustive authority total is unresolved.

UNKNOWN: Full distribution, LP custody/lock coverage, issuer authentication, real freshness and the cause of the unresolved R4/R2 freeze-authority conflict.

INSUFFICIENT EVIDENCE: “Official and safe, all liquidity locked, freeze impossible” fails against absent or conflicting evidence. Repeated F-A records are not independent sources.

Using generation time as observation time; inventing units for T+ labels; ranking missing inventory as one holder; treating multiple provider fields as independent confirmation; resolving old/new conflicts silently; and issuing predictions from a diagnostic report.

Prepare a five-part report: scope, evidence register, four-state findings, limitations and three next checks. Calculate displayed quantities and coverage. Correct the supplied headline “Official safe token: all liquidity locked and freeze impossible.” Include R4 without contaminating S0 with older state.

• State object and cutoff before the conclusion.
• Separate observation, capture and generation context.
• Bind each claim to records and assumptions.
• Publish coverage and denominators.
• Preserve conflicts and unavailable fields.
• Request missing evidence without a trading call.
• Keep needs_review and version truthful.

A defensible asset report makes its evidence chain recoverable. Narrow supported findings remain useful alongside unresolved identity, authority or liquidity questions.

Provenance is part of the claim. Partial coverage must remain partial. Diagnostic records preserve observations and limits; they do not certify safety or predict returns.

Create evidence register R1–R6 with source/context columns and shared F-A bracket. Four-state report panel links claims to rows. Show50% coverage with unreturned composition unknown; S−2 and S0 separate. Label T+ values synthetic without seconds/minutes. Preserve R4/R2 as conflicting assertions: label cause UNKNOWN / UNRESOLVED and show no reactivation arrow. Cite S08-SET/S08-FREEZE for permanent revocation semantics only; do not authenticate fixture values.

Illustrative diagnostic report; neither a runtime Passport nor a safety certificate.

Use deterministic SVG/HTML or charts with text alternatives, dark navy/black and restrained cyan, electric blue and violet. Label every quantity and uncertainty. At 390px stack panels; 768px/desktop and Arabic RTL verification remain future asset work. Keep identifiers LTR and factual axes/edge directions unchanged. Use only the official supplied ZECOIN mark if branding is added. Both records are specifications, not delivered assets.

Use DIAGNOSTIC_PASSPORT for the lesson's bounded educational purpose only if network support, exact identity, fields and provenance are verified. No runtime endpoint or provider capability is asserted by this content. Use the embedded offline fixture without paid access. Missing data stays unknown; no wallet connection, transaction signature, token launch or live trade is required. Academy progress, payments and referrals never change Radar evidence. A Diagnostic Passport preserves observations and limits, not a price prediction or safety certificate.

• Solana getTokenSupply — Raw supply, decimals and RPC response context; not a circulating-supply definition. Checked 2026-09-30.

• Metaplex Token Metadata Updating Assets — Update authority, mutable flag and editable metadata fields within this specific program. Checked 2026-09-30.

• Uniswap v2 Pools — Fungible LP contribution and redemption claims, specific to v2. Checked 2026-09-30.

• Solana getTokenAccountsByOwner — Owner-scoped enumeration filtered by mint or program. Checked 2026-09-30.

• Solana Set Authority — A valid revocation to None permanently removes the selected authority role; scoped here to the same correctly identified standard Token Program mint and freeze role. Mechanism only, not fixture authentication. Checked 2026-10-01.

• Solana Freeze Account — Only a mint with freeze authority can freeze token accounts; revoked freeze authority cannot be used to freeze accounts. Mechanism only, not fixture authentication. Checked 2026-10-01.

Documentation supports mechanisms, not fictional dataset values. All examples and exercise records are original ILLUSTRATIVE material, frozen in this file. S0/S1 are synthetic context labels; observedAt is null because no live or historical ledger measurement was collected. Re-review documentation and actual deployed behavior before any publication or integration.

P08 is complete. Review your source-to-claim report and unresolved limitations, then return to the Academy dashboard. Later programs and separate labs retain their listed prerequisites.