298 lines
10 KiB
Markdown
298 lines
10 KiB
Markdown
# Krea2 Prompt Guide
|
|
|
|
This document records prompt rules discovered from actual SxCP generator
|
|
outputs tested in Krea2. It is not a generic prompt cookbook. Add a rule only
|
|
when an A/B image comparison shows that the wording improves or breaks Krea2
|
|
behavior.
|
|
|
|
## Core Rule
|
|
|
|
Krea2 responds best when the prompt gives one clear visual hierarchy:
|
|
|
|
1. subject/cast descriptor,
|
|
2. action or pose,
|
|
3. clothing state,
|
|
4. location,
|
|
5. camera/layout,
|
|
6. expression,
|
|
7. composition/crop,
|
|
8. style.
|
|
|
|
Avoid letting two sections describe incompatible camera or framing intents.
|
|
|
|
## Prompt Output Contract
|
|
|
|
- `sxcp_eval_out` must contain only the prompt being tested.
|
|
- Analysis, scoring, and generator notes belong in chat or `sxcp_eval_log`.
|
|
- Keep one experiment variable per cycle when possible.
|
|
- Lock seed, character, location, and camera when testing wording changes.
|
|
- Treat the MCP seed as transport metadata. Preserve it for prompt-only A/B tests
|
|
and do not write it into the visible prompt text.
|
|
|
|
## Seed-Controlled A/B Tests
|
|
|
|
Use one fixed seed when deciding whether prompt wording helped Krea2. A single
|
|
image can justify a prompt-only retry when the mismatch is obvious, but a
|
|
generator rule needs either repeated evidence or a generated prompt that is
|
|
structurally wrong before rendering.
|
|
|
|
When a workflow batches soft/hard prompts through an index switch, sidecar text
|
|
files may not be the exact prompt used for each rendered image. If the sidecar
|
|
and image disagree, inspect the PNG workflow metadata and the final text encode
|
|
input before patching the generator.
|
|
|
|
When reviewing an eval payload, log:
|
|
|
|
- emitted seed,
|
|
- original generated prompt,
|
|
- edited prompt,
|
|
- image failure or improvement,
|
|
- whether the change should stay prompt-only or become a generator patch.
|
|
|
|
## Camera And Composition
|
|
|
|
### Orbit / Multiangle Camera
|
|
|
|
When Krea2 receives an orbit or multiangle camera, avoid selfie-specific wording
|
|
unless the intended camera is actually a handheld or mirror selfie.
|
|
|
|
Works better:
|
|
|
|
- `lifestyle portrait frame`
|
|
- `creator portrait frame`
|
|
- `outfit-check pose`
|
|
- `wide environmental coworking camera layout`
|
|
- `camera placed several meters away`
|
|
- `full seated body from head to knees`
|
|
- `room depth surrounding the subject`
|
|
|
|
Conflicting wording:
|
|
|
|
- `selfie frame`
|
|
- `phone selfie`
|
|
- `holding the phone`
|
|
- `creator-shot phone photo`
|
|
- `handheld camera realism`
|
|
|
|
Observed result: selfie words pulled a back-right elevated wide shot into an
|
|
arm-length selfie. Removing selfie terms made the image follow the rear-quarter
|
|
view much better.
|
|
|
|
### Wide Shots
|
|
|
|
Krea2 tends to keep attractive subjects large in frame. To get a real wide or
|
|
environmental frame, be explicit about distance and visible environment.
|
|
|
|
Useful phrasing:
|
|
|
|
- `camera placed several meters away across the desk aisle`
|
|
- `full seated body from head to knees remains visible`
|
|
- `nearby desk edge, laptop corner, repeated desk rows, and tall-window depth clearly readable`
|
|
- `wide environmental room framing`
|
|
|
|
Avoid relying on `wide shot` alone.
|
|
|
|
## Location Layout
|
|
|
|
Location-aware camera text works when it describes the room around the subject
|
|
without stealing the foreground from the subject.
|
|
|
|
For coworking lounge:
|
|
|
|
- Keep `warm desks`, `laptop tables`, `glass partition seams`, `repeated desk rows`,
|
|
`plants`, and `tall windows`.
|
|
- Mention foreground anchors only when the camera should actually see them.
|
|
- In POV, keep location anchors beside or behind the bodies, not in the lower
|
|
foreground.
|
|
|
|
## Clothing Continuity
|
|
|
|
When a softcore outfit is reused in a later branch, name what happens to actual
|
|
outfit pieces instead of using generic fabric language.
|
|
|
|
Works better:
|
|
|
|
- `denim shorts are pulled aside or removed below the hips`
|
|
- `button-down shirt tied at the waist and fitted bralette remain visible from the same outfit`
|
|
|
|
Avoid generic fallback wording:
|
|
|
|
- `fabric slipping off`
|
|
- `partly exposed`
|
|
- `outfit pushed aside where needed`
|
|
|
|
Use generic wording only when no source outfit exists.
|
|
|
|
## Climax / Ejaculation Wording
|
|
|
|
Krea2 needs the visible fluid action stated directly in climax routes. Keep
|
|
`semen` in the action sentence when the generator intends an ejaculation or
|
|
post-ejaculation image.
|
|
|
|
Works better:
|
|
|
|
- `the man ejaculates semen across her body`
|
|
- `visible semen lands across her body`
|
|
- `ejaculates semen across her lower back, ass, and thighs`
|
|
|
|
Avoid relying only on softer aftermath wording:
|
|
|
|
- `visible ejaculation`
|
|
- `visible orgasm aftermath`
|
|
- `cum on hands and body`
|
|
|
|
Seed `42052` exposed this as a structural formatter issue: the route selected a
|
|
standing/kneeling climax pose, but the Krea prompt dropped `semen` and kept only
|
|
generic aftermath language. The generator now preserves direct `ejaculates
|
|
semen` wording for this path before the prompt reaches Krea2.
|
|
|
|
## POV Outercourse
|
|
|
|
### Boobjob / Titjob
|
|
|
|
The atlas examples are frontal and upright: the visible partner faces the viewer,
|
|
kneels between the viewer's thighs, and compresses the shaft between the breasts.
|
|
Forward-bent wording can still place the body correctly, but it weakens the
|
|
breast contact.
|
|
|
|
Works better:
|
|
|
|
- `POV boobjob position`
|
|
- `woman kneels upright between his legs facing him`
|
|
- `penis rises vertically in the lower foreground`
|
|
- `squeezed between her pressed-together breasts`
|
|
- `woman's own fingers and nails cup her breasts from the outside`
|
|
- `glans emerging above the cleavage directly below her mouth`
|
|
|
|
Avoid vague or conflicting wording:
|
|
|
|
- `torso bent forward over his pelvis`
|
|
- `both hands push her breasts` without naming whose hands
|
|
- `only foreground hands` when the intended hands are the woman's hands
|
|
|
|
### Handjob
|
|
|
|
The atlas examples are direct and simple: the woman faces the viewer between his
|
|
thighs, her hand is the visible contact point, and her face or torso stays behind
|
|
the shaft. In fixed-seed tests, naming `one hand` plus a generic POV camera line
|
|
could make the lower hand read as the viewer's hand. Naming the woman's hands
|
|
made the action clearer.
|
|
|
|
Works better:
|
|
|
|
- `POV handjob position`
|
|
- `woman kneels between his legs facing him`
|
|
- `the woman's right hand wraps around the viewer's penis`
|
|
- `her left hand steadies the base`
|
|
- `viewer thighs and pelvis frame the lower edges`
|
|
- `without his hands covering the action`
|
|
|
|
Avoid:
|
|
|
|
- generic `one hand grips` when hand ownership matters
|
|
- allowing `foreground hands` to compete with the woman's active hand
|
|
|
|
## POV
|
|
|
|
In POV prompts, the visible subject should still be established first. The POV
|
|
participant is the camera viewpoint, not a normal visible cast member.
|
|
|
|
Works better:
|
|
|
|
- visible subject descriptor first,
|
|
- then POV action,
|
|
- then foreground hands/body/clothing cues.
|
|
|
|
For POV clothing, describe only visible body/clothing fragments:
|
|
|
|
- `foreground hands, hips, thighs, or lowered waistband`
|
|
- `foreground hands, forearms, sleeves, or torso edge`
|
|
|
|
Avoid:
|
|
|
|
- full third-person `Man A wears...` phrasing for the POV participant,
|
|
- making `the viewer` the first subject before the visible character is
|
|
established.
|
|
|
|
For POV climax wording, the fluid target must follow the pose before expression
|
|
tokens. Rear-entry, doggy, bent-over, face-down, and on-all-fours poses should
|
|
target the ass, thighs, and lower back even if the expression detail mentions
|
|
face, lips, mouth, or tongue.
|
|
|
|
Evidence:
|
|
|
|
- Dataset seed `52` generated an internally contradictory prompt: on-all-fours
|
|
rear-view positioning paired with a face/chest ejaculation target.
|
|
- Corrected seed `52` and follow-up seed `5202` both rendered the rear-view
|
|
target consistently when the wording used `across her ass, thighs, and lower
|
|
back` and kept the clothing state tied to the lower garment.
|
|
|
|
### POV Doggy / Rear-Entry
|
|
|
|
For doggy-style POV, visible viewer thighs, lower torso, or pelvis can be
|
|
correct. Real POV references often show them. The useful target is not removing
|
|
the viewer body, but making the body cues read as a standing or crouched
|
|
first-person viewpoint instead of a vague seated pose.
|
|
|
|
To push the reference closer to a standing or crouched man looking down, use a
|
|
top-down rear-entry structure:
|
|
|
|
- `top-down standing POV doggy position from behind`
|
|
- `camera looks down over the viewer's extended hands onto the woman's raised hips`
|
|
- `woman is on all fours with chest low, forearms folded, cheek turned sideways`
|
|
- `rear-entry penetration visible between raised hips`
|
|
- `face and mouth remain far ahead, clearly separated from the penis`
|
|
|
|
Do not use visible shoes or lower legs as the standing cue. Seed `65` showed
|
|
that adding shoes/lower legs made Krea2 drift into oral contact and lose the
|
|
rear-entry geometry.
|
|
|
|
Do not over-prompt `viewer torso and thighs outside frame`; seeds `65` and
|
|
`6602` showed Krea2 still draws lower-body POV cues, and real references support
|
|
that. Prefer framing them as plausible foreground body cues rather than trying
|
|
to suppress them.
|
|
|
|
## Stronger-Control / Low-Priority Cases
|
|
|
|
Some atlas routes are useful to catalog but are not good prompt-only tuning
|
|
targets yet. Keep these out of the normal fixed-seed prompt queue until easier
|
|
pose families are covered.
|
|
|
|
### Sixty-Nine / Close Reversed POV
|
|
|
|
`pov_sixty_nine_close_reversed_oral` is currently the hardest and low-priority
|
|
atlas route. It should be treated as a pose/control image or image-guided route
|
|
first, not a normal prompt-only fixed-seed candidate.
|
|
|
|
The repeated atlas geometry is close and specific: the visible partner is
|
|
reversed over the viewer, hips closest to the camera, head and torso receding
|
|
away into the upper frame, and the viewer face or mouth anchoring the lower
|
|
foreground. Text-only prompting can collapse this into generic oral contact or
|
|
lose the reversed-over-viewer body arrangement.
|
|
|
|
## Style
|
|
|
|
Style should describe rendering, not camera mechanics.
|
|
|
|
Use style presets to choose between:
|
|
|
|
- natural photo,
|
|
- creator/social-media photo,
|
|
- documentary/direct-flash photo,
|
|
- cinematic realism,
|
|
- illustration/comic.
|
|
|
|
If a controlled camera is active, avoid style suffixes that imply a conflicting
|
|
camera such as `phone photo` or `handheld selfie`.
|
|
|
|
## Guide Update Format
|
|
|
|
When adding a new rule, include:
|
|
|
|
- observed prompt,
|
|
- observed image failure,
|
|
- edited prompt wording,
|
|
- image improvement or regression,
|
|
- generator path if known,
|
|
- final rule.
|