| name | hima-survey |
| description | Map the problem before any charter exists — topology, dependencies, unknowns, requirement framing. |
| stage | discovery |
| source | base |
| mode | force |
Purpose
Cover the cycle's discovery + analysis stages: build a shared map of the problem before a
single line of charter or code exists. hima-survey exists so that "what are we actually solving"
is answered on the record, not assumed. At M+ criticality it is a non-negotiable gate — no
charter may open without a recorded survey.
When to use
- A new feature, bug, or initiative enters the cycle and no survey artifact exists yet for it.
- The problem topology is unclear: unknown dependencies, ambiguous scope, or competing framings
of "what this is."
- At T/L criticality this skill is advisory — a lightweight note is enough; do not over-invest.
Steps
- Restate the problem in the requester's own language before reframing it technically.
- Map topology: what files/systems/data this touches, what already exists, what does not.
- Enumerate dependencies and unknowns explicitly — do not silently assume an answer.
- Frame candidate requirements as testable statements (inputs → outputs → constraints).
- Flag anything that looks like a true blocker for
hima-lookout or hima-parley, not for
silent assumption.
- Record the survey as a durable artifact the next stage (
hima-charter) can read.
Done when
- The problem topology, dependencies, and unknowns are written down, not just discussed.
- Candidate requirements exist in a form
hima-charter can turn directly into Gherkin scenarios.
- No unresolved true blocker remains unflagged.