steering-conventionslisted

# Data Conventions If vault_path is not in context, read `~/.myna/config.yaml` first. If the file does not exist, tell the user to run `/myna:setup` and stop. ## Provenance Markers Every agent-written entry carries exactly one marker: | Marker | Meaning | Write path | |--------|---------|------------| | `[User]` | User typed it directly | Direct write | | `[Auto]` | All data explicit in source | Direct write | | `[Inferred]` | Some fields are agent's guess | Direct write, flagged | | `[Verified]` | Was Auto/Inferred, user confirmed | After review queue approval | **Decision framework:** | Signal | Tag | |--------|-----| | User typed it | `[User]` — always. User made the judgment. | | All data explicit in source (names, dates, actions stated) | `[Auto]` — can reconstruct from source without guesswork. | | Core item is real but some fields guessed | `[Inferred]` — write it, flag it. User verifies when they notice. | | Multiple valid interpretations | Review queue — two reasonable people would read it differently. Don't guess. | **Litmus test for `[Inferred]` vs review queue:** Is there one obvious-enough interpretation? `[Inferred]`. Would two reasonable people read it differently? Review queue. **`[Inferred]` highlighting:** Features that compile data (brag docs, performance narratives, self-reviews, person briefings) must flag `[Inferred]` entries so the user knows which data points to double-check. ## Date + Source Format Every vault entry includes a date and sourc