A lesson photo can show a space or material without establishing every claimed outcome. Reading the actual craft-work asset and post mapping led me to examine how a portfolio description obtains its evidence.
Moving an image also moves an explanation
Rebuilding a portfolio can move image files, posts and card copy through different paths. The source work records asset-to-post and evidence relationships. I read that as more than a file inventory: it can retain which sentence relies on which asset. A filename alone does not establish the claim’s scope.
No actual lesson photo, person or location appears here. The independent example uses one original shape and the narrow claim that it is visible. This exposes the production decision without exposing original media or customer activity records. The relationship contract remains explainable without copying those assets.
Build a reference from asset to claim
Lineage accepts asset identifiers, origin and permission notes, then links claim identifiers, asset references, statements, evidence and caller review. Orphan claims, repeated identifiers and missing true review flags fail. These checks identify which file and sentence should be examined together.
A true flag does not prove what someone actually reviewed. The code performs neither image recognition nor consent verification. Reference integrity and judgment about a suitable description are separate responsibilities; output explicitly avoids claiming automatic truth assessment.
Distinguish visible facts from inferred outcomes
A visible synthetic shape and a claim about what someone learned from it require different evidence. A photo works similarly. Visible space and tools should not automatically become proof of satisfaction or educational effect. This example introduces no actual learning-effect claim.
A caption can distinguish visible content from context separately confirmed by its provider. Reading source work also requires distinguishing requested wording from wording supported by the material. Plans and final files may differ, so public prose should retain only the scope actually checked.
Recheck claims when an image changes
Overwriting another file under the same identifier could escape reference-only checks. This small module implements no byte hash or asset store. An adopter should add file versioning or hashes and re-review connected claims after replacement; a stable identifier cannot guarantee identical evidence.
A permission note is also not a grant of rights. Actual publication permission and personal-information handling need separate verification. I retain these limits to distinguish automated integrity from human review instead of pretending that importing more customer photos would resolve them.
Connect portfolio persuasion to evidence
The package’s twelve tests include linked references, orphan claims, duplicate assets and preserved inputs. Its viewer reads synthetic results from actual local runs, without loading customer photos or changing the original site. Only new reusable code is packaged.
A portfolio should explain why an image was chosen and what it helps describe. I want the selection decision and its evidence to remain visible together. Other industries can use the same distinction between what an image shows and what the adjacent sentence needs to establish, producing concrete stories without exaggerated outcomes.
Handle a missing basis without borrowing visual authority
An attractive photo should not supply borrowed credibility to an unsupported sentence. The maker can narrow wording to the observed scene or record the need for further material. The orphan-reference error follows that direction: missing evidence is not replaced with a plausible default. This helps the next editor distinguish a genuinely supported description from one that still needs work, without inventing an outcome to make a portfolio appear stronger.