The StandIn Protocol

How StandIn draws the line

We do not sell "magic" AI that knows everything.
We sell continuity software that respects privacy.

Security Boundary Statement

StandIn's protocol is enforced technically, not socially.

The system cannot access private messages, drafts, or unpublished content by design and by permission scope. There is no configuration, admin setting, or escalation path that enables this access.

If content is not explicitly published into a Wrap, committed to a system of record, or linked as a public artifact, it is invisible to StandIn.

This boundary is enforced automatically and cannot be bypassed.

The Response Spectrum

Every interaction falls into one of three strict categories.

1. We Answer

Explicitly Published

If a human wrote it in a Wrap, committed it to code, or updated a linked ticket, StandIn treats it as fact.

Only when the source is explicit, published, and within scope.
If any of these conditions are missing, StandIn will not answer.

Query
"Did Sarah deploy the fix?"
Answer
"Yes. Sarah noted in her wrap that PR #402 was deployed to staging at 17:00."

2. We Redirect

Unknown but Routable

If the answer isn't in the record, StandIn identifies who holds the context and points you to them.

Redirection does not expose private context.
It only identifies an owner already listed in the declared record.

Query
"Why is the build failing?"
Response
"There is no incident report in the wrap. Dave is listed as the on-call engineer."

3. We Refuse

Private or Speculative

If it requires guessing, reading private messages, or inferring sentiment, StandIn stays silent.

Refusal is enforced at the system level.
Prompt phrasing, rewording, or escalation cannot bypass it.

Query
"Is Sarah annoyed?"
Refusal
"I cannot answer that. I do not track personal sentiment or private conversations."

The Prohibited List

These constraints are enforced by design. They are not feature toggles.

These are hard-coded constraints. There is no "admin override" for these features because they do not exist.

No Passive Monitoring

We do not watch your screen, track your mouse, or log your active hours.

No Private Message Ingestion

DMs are black boxes to StandIn. If you didn't post it publicly, it doesn't exist.

No Intent Inference

We do not guess why someone did something. We only report what they explicitly said.

No Management Oversight

Managers cannot query StandIn for "who is working hard?" or "summarize activity".

StandIn does not log, analyze, or summarize behavior in order to reconstruct these prohibited signals indirectly.

Enforcement Over Intent

StandIn does not rely on user intent or policy compliance to stay safe.

Fixed data scopes

Technical limits on data access that prevent scope creep.

Role-specific representatives

Agents are restricted to their defined domain and cannot access other data.

Time-bounded authorization

Access grants expire automatically and must be renewed by human action.

Hard refusal rules

The system is hard-coded to refuse privacy-violating queries.

Even well-intentioned misuse is blocked by default.

There is no override

Safe enough for your team?