Write and audit the lossless-codec card

The final card explains one bounded byte codec. Its confidence comes from exact identity, state, recovery, and denominator evidence, not from a familiar name.

Goal

Write and audit project-root report.md as a concise lossless-codec card whose claims point to saved artifacts and respect every interpretation limit.

Inputs

Use the manifest, source inventory, encoder and decoder traces, dictionary summary, document codes and digests, payload/container records, malformed results, round-trip table, size accounting and totals, limit audit, figures and plot data, read-back checks, and replay agreement or mismatch.

Deliverables

The card must explain in plain language:

• source paths, bytes, strict UTF-8 text, digests, and identity boundaries;
• the fixed 256-entry byte dictionary, per-document reset, bounded growth, longest-phrase encoder, and special next-code decoder rule;
• two-byte unsigned big-endian codes, exact DTC1 payload fields, and FILE-15 outer container fields and failure order;
• malformed-code and malformed-payload behavior, recovered length, bytes, digest, text, code, and final-size agreement;
• body, payload, and container size denominators, empty null ratios, corpus aggregation, and dictionary-limit audit effects;
• artifacts, read-back, figures, exact plot data, replay result, and first mismatch behavior; and
• what the fixture cannot establish about standard compatibility, universal compression, optimality, corruption detection, or speed.

Include one small worked path from source bytes through dictionary additions, codes, packed fields, decoding, and exact recovery. Distinguish observations from interpretation and name any first mismatch honestly.

Checks

Audit every statement against a saved record or explicit rule. Reject wording that claims standard LZW compatibility, character-symbol operation, universal size reduction, optimality, production readiness, general corruption detection, or unmeasured speed. Check that all ratios name their denominator, empty ratios remain null, and header overhead remains visible.

Check the card names special-code and bound behavior, fixed width, malformed streams, strict UTF-8 recovery, digest equality, primary and audit limits, artifact read-back, and replay outcome. A reader should be able to locate the evidence behind every numerical claim.

Workspace

Keep the card at the project root as report.md. Do not replace it with a shrinking-file animation or percentage gauge. Preserve machine-readable artifacts, exact bytes, digests, and figures beside it.

Hints

Review

Give the card to someone who has not opened the source code. Can they explain the special code, identify header overhead, find a malformed-stream result, reproduce one ratio, and tell whether replay agreed? Mark every sentence that outruns the artifacts.

How to check your work

Checks compare report.md with the evidence-limited reference card after your own audit. The supplied fixture explains the exact byte contract and measurements without claiming standard compatibility or universal compression.