AI Import Column Mapping
Guess how an uploaded file's columns line up with the app's fields, then ask.
What it adds
A proposed mapping from uploaded columns to destination fields, shown with samples and confirmed before the import runs.
What your agent is told to do
5
What your agent is told to do
5-
1
Take the uploaded headers and a small sample of rows and propose a mapping to the destination fields. Match on exact and near-exact header names first and only consult the model for the columns that remain unresolved.
-
2
Constrain the candidate set to fields the current user and the importer are actually permitted to write. A mapping that targets a read-only, computed, or privileged field must never be offered, whatever the header says.
-
3
Show the proposal as a table: source column, sample values, destination field, confidence, and the transformation that will be applied. A user cannot approve a mapping they cannot see the effect of.
-
4
Require an explicit confirmation before any row is written, with low-confidence and unmapped columns called out for attention. Do not start the import on the strength of the suggestion alone.
-
5
This brief ends when the mapping is confirmed. Rows that then fail validation belong to AI Import Error Repair, and corrections to data already stored belong to AI Data Cleanup Suggestions; hand off rather than duplicating their review screens.
Edge cases it handles
8
Edge cases it handles
8- Only fields the importer may write can be offered as targets. Filter the candidate list by permission before the suggestion is made, not after the user has chosen.
- Every proposed mapping needs a confidence and a worked sample showing what a real value becomes after transformation, so a wrong date or currency assumption is visible before the import rather than after.
- Two source columns silently pointing at one field will overwrite each other unpredictably. Detect the collision, refuse to proceed, and make the user resolve it.
- Custom fields defined per workspace and headers in another language both need to map. Read the destination list at runtime rather than assuming a fixed schema, and match on meaning rather than English keywords alone.
- No row is written until the user confirms. An import that begins while the mapping preview is still on screen cannot be taken back.
- Columns the user leaves unmapped must be reported explicitly before confirmation, because silently dropped data is discovered weeks later.
- Files with no header row, duplicate headers, or a header row that is actually data must be handled rather than mapped confidently to nonsense.
- Send only headers and a bounded sample of rows to the model, never the whole file, and set a ceiling on that sample so a large upload cannot run up an unbounded cost.
Definition of done
9
Definition of done
9- Uploaded columns arrive with a proposed destination field, a confidence, and a sample transformation.
- Only fields the importer and the current user may write appear as mapping targets.
- Two columns mapped to the same field block confirmation until resolved.
- Unmapped columns are listed explicitly before the user confirms.
- No rows are written until the user confirms the mapping.
- Workspace-specific custom fields and non-English headers map correctly.
- Only headers and a bounded row sample are sent to the model.
- The feature matches the existing design system.
- No existing functionality is broken.
Related features
Prompt Injection Defense
Prompt Injection Defense
Stop instructions hidden in documents, pages, and tool output from steering the AI.
What it does
A boundary between the app's own instructions and untrusted content, backed by enforced tool permissions and confirmation for consequential actions.
How it works
- 1 Enumerate every route by which content the app did not author reaches a model: uploaded files, retrieved passages, fetched pages, records from connected accounts, and the results of tool calls.
- 2 Keep the app's own rules in the trusted portion of the request and enclose untrusted content in a clearly delimited region labelled as material to be analysed rather than instructions to be obeyed.
- 3 Enforce what the model may reach outside the model. Tool permissions, tenant scoping, and rate limits must hold even when the model has been completely persuaded — an instruction telling the model not to delete things is not a control.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/prompt-injection-defense
Model Fallback
Model Fallback
Keep AI features working when the primary model provider fails.
What it does
A bounded retry path that reruns a failed AI task on a compatible alternative model and discloses the substitution.
How it works
- 1 Classify failures before reacting. Provider outages, capacity rejections, rate limiting, and timeouts are worth retrying elsewhere; a refusal, an invalid request, or an authentication error will fail identically on any destination and must surface immediately.
- 2 Honour rate limiting with backoff before moving on. Retrying instantly against a provider that asked you to wait deepens the outage and can extend the penalty period.
- 3 Allow fallback only to models that can honour the same output shape and the same tool behaviour as the original. A substitute that returns a different structure turns a provider outage into a data problem inside the app.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/model-fallback
AI Audit Trail
AI Audit Trail
Keep a reviewable record of what the AI saw, produced, recommended, and actually did.
What it does
A per-run record of inputs, sources, configuration, output, approvals, and executed actions, restricted and retained by purpose.
How it works
- 1 Write one record per run covering the feature invoked, the configuration and prompt version in force, the sources retrieved, the tools called, the output produced, who approved it, and what was carried out.
- 2 Distinguish a draft from an executed action explicitly in the record. A trail that shows only what text was produced cannot answer the question that matters in a review, which is whether anything happened.
- 3 Redact at the moment of writing. Strip credentials, tokens, and payment details before the record is stored, and keep only the personal content genuinely needed to understand the decision.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/ai-audit-trail
How it works
-
1
Copy the link
Grab the Markdown instruction URL for this feature.
-
2
Give it to your AI
Paste it into Claude Code, Cursor, v0, Lovable — whatever you build with.
-
3
It inspects, then implements
Your agent reads your existing app first, then adds the feature to fit it.
Works with your stack
These instructions are written to adapt. They tell the agent to detect your framework, match your existing design system, and reuse what you already have — rather than assuming a particular stack.
Need it tighter than that? Customize the feature and tell it exactly what you're running.