Field Protocols

Runnable instruments. Receipts, no theater.

Field Protocols are short, executable instruments for keeping reality intact when systems and organizations get noisy. They are designed to work on bad weeks, under pressure, and with imperfect information.

Every protocol is written to be:

specific enough to run
strict enough to verify
small enough to ship

Core wrapper: OTW (Owner, Timebox, Witness)
A protocol is not “done” when it is said. It is done when it is owned, time-bounded, witnessed, and verified.

OTW (Owner, Timebox, Witness)
Owner: the accountable human for the next delta.
Timebox: the maximum time you will spend before shipping or stopping.
Witness: what observes completion (a person, a channel, a log, a calendar).
Evidence: what exists afterward that survives scrutiny (a link, screenshot, export, diff, metric delta).

MPTT (Map, Probe, Trace, Teach)
Map
Name the workflow class, decision rights, and the reversal path. List artifacts that must exist for reconstruction (logs, approvals, configs). Declare the failure boundary: what must never happen, and who owns that boundary.

Probe
Run one small test that creates a decision point (exception request, rollback drill, access revoke). Time the decision and capture what broke first. Record the first disagreement, reversal, or rework loop as telemetry, not drama.

Trace
Capture the decision chain: request, authority, approval, execution, evidence. Score it with one admissible proof metric (not vanity). Store only evidence classes if confidentiality applies (what exists, not the content).

Teach
Convert the trace into a reusable instrument within 14 days: checklist, rubric, or gate. Publish the instrument and its proof linkage in the MCRI (Master Catalog and Routing Index). Repeat with ruthless consistency until the primitives become default handles.

How to use this library

Pick one protocol.
Run it once inside the timebox.
Post the OTW line and the evidence link.
If you miss, post the reason and a new time. Do not go silent.