SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsdreamina-cli-image2image

dreamina-cli-image2image

Design & creation

View repository

Updated 10-15 21:47

About this skill

Jimeng CLI image-to-image

⚠️ Maintenance discontinuation notice: The legacy Dreamina CLI (dreamina, v1.4.18) will be November 2026 starting from which maintenance will cease. For new tasks, switch to the Canvas replacement skill dreamina-canvas-cli-image2image(installation:npx skills add full-aigc-skills/dreamina-skills --skill dreamina-canvas-cli-image2image). The content of this skill is retained as a historical contract reference.

Before execution, run dreamina image2image -h. This skill records the v1.4.18 stable workflow; the actual help is always the source of truth for parameters.

When to use and boundary

For tasks that already have local reference images and need editing and result tracking via the CLI. It should not be used for text-to-image, video generation, or pure prompt polishing; load the corresponding execution or prompt skill respectively.

Collaboration routing:dreamina-prompt-image2image is responsible for editing prompts; this Skill is responsible for local dreamina CLI execution; when the user explicitly selects OpenCLI transport, switch to dreamina-opencli-image2image, do not submit repeatedly.

Must comply

  • --images accepts 1–10 local files, multiple paths separated by commas.
  • Each submission explicitly passes --resolution_type=1k|2k|4k。
  • --width and --height must appear in pairs, be positive integers, and be mutually exclusive with --ratio .
  • Only 5.0Pro supports 1k; 4.x/5.0 only support 2k or 4k.
  • The model token is 5.0Pro, not the display name 5.0 Pro。
  • generate_num range is 1–10.

For the complete model matrix and pixel limits, see dreamina-cli skill's references/dreamina-cli-v1.4.18-contract.md. If not installed, first npx skills add full-aigc-skills/dreamina-skills --skill dreamina-cli。

Standard execution

test -r ./input.png
dreamina user_credit
dreamina image2image \
  --images=./input.png \
  --prompt="保留主体与构图，转换为透明水彩风格" \
  --model_version=5.0 \
  --resolution_type=2k \
  --ratio=1:1 \
  --poll=0

Omit ratio when using custom dimensions:

dreamina image2image \
  --images=./input.png \
  --prompt="改造成竖版电商海报" \
  --model_version=5.0 \
  --resolution_type=2k \
  --width=1536 \
  --height=2048 \
  --poll=0

Image upscaling

Image upscaling is routed by this Skill, but the real-time help must be read first, because the parameters differ from image-to-image:

dreamina image_upscale -h
dreamina user_credit
dreamina image_upscale \
  --image=./input.png \
  --resolution_type=2k \
  --poll=0

Before submission, verify that the source images are readable, disclose the resolution and consumption impact, and obtain explicit authorization. Save the returned submit_id, via dreamina query_result track until success or fail。

Asynchronous closed loop

Step 1: Verify

Verify one by one that the input files are readable, and use help to validate the model, resolution, ratio, or custom dimension combination.

Step 2: Submit

Check credits, explain the consumption impact, submit, and save submit_id。

Step 3: Final-state closed loop

For querying continuously call query_result, until success or fail. On failure, report fail_reason, do not treat submit success as generation success.

Gotchas

  1. Number of files: only 1–10 readable local images are allowed.
  2. Missing resolution: v1.4.18 must pass resolution_type。
  3. Width-height conflict: width/height come in pairs and are mutually exclusive with ratio.
  4. 1k misuse: only 5.0Pro supports image-to-image 1k.
  5. Status misjudgment：querying is not a successful final state.

References

When to use (when to use)

When the user needs Execute generation or write operations after clarifying inputs, budget, and delivery constraints load this skill. First extract the goal, inputs, constraints, delivery format, and acceptance criteria from the request; the description summary is: Use when the user provides local images and wants Dreamina 即梦 image-to-image editing or image upscaling through dreamina image2image or dreamina image_upscale, including validation, submission, polling, and troubleshooting.。

Rules

  • Read before write: first confirm the current state and actual capabilities, then perform actions that change external state.
  • Least privilege: use only the files, tools, accounts, and network scope required to complete the current step.
  • Evidence first: when run results, resource IDs, versions, hashes, or test outputs are missing, clearly mark them as NOT_VERIFIED。
  • Idempotency first: retain request identifiers and phase status; when results are unclear, query first and do not retry blindly.
  • Privacy and security: logs, examples, receipts, and error messages must not contain tokens, cookies, keys, or personally sensitive data.

Workflow

Step 1: Clarify intent

Confirm whether this skill matches the goal; if it is only an adjacent need, hand it off to a more precise skill.

Step 2: Perform preflight checks

Validate inputs, model/tool capabilities, output paths, budget limits, and approval status; if any critical condition is unknown, stop at the read-only stage.

Step 3: Form a plan

List the tools to be called, the objects that will change, the success criteria, and the safe exit method after failure.

Step 4: Execute actions

Execute once as approved and record the request identifier; for ambiguous results, query first instead of resubmitting; retain an associable status or receipt for each external call.

Step 5: Verify delivery

Verify artifact existence, format, hash/identifier, cost status, and quality gates, and state facts, inferences, and unverified items separately.

Validation checklist

  • The skill trigger conditions match the user's intent, and adjacent tasks were not misrouted to this skill.
  • Inputs, target objects, versions, and output locations are all clear, and no guessed values were used in place of required values.
  • All write, paid, publish, or irreversible actions are within the scope of user authorization.
  • Results have been verified with independent checks; "command succeeded" or "file exists" alone does not count as complete acceptance.
  • The output includes actual evidence, failed/skipped items, remaining risks, and executable next steps.

Gotchas

  1. Mistaking the plan for the result: a document or prompt is not equal to actual execution; the actual execution level must be indicated.
  2. Retrying on error: a timeout or lost response may already have produced remote state; query first before deciding whether to retry.
  3. Implicit scope expansion: batch, full-volume, publish, overwrite, and paid operations are not natural extensions of ordinary reads and writes.
  4. Version drift: when referencing external resources, record the version, tag, or commit; do not treat a mutable branch as release evidence.
  5. Stale evidence: caches, old screenshots, and historical tests cannot prove the current environment; refresh key evidence before delivery.

Non-applicability and boundaries

Paid, publish, overwrite, upload, or external write operations must use the explicit authorization of the current task; do not automatically expand the number of attempts or the budget. If the request requires another skill, do not copy its body text; hand off by skill name and preserve the current task context.

Progressive disclosure

  • When inputs/outputs, status, and authorization points need to be determined, read references/workflow-contract.md。
  • When pre-delivery self-checks are needed, read references/validation-checklist.md。
  • When encountering timeout, partial success, or recovery scenarios, read references/error-recovery.md。
  • For first run, privilege escalation denial, and failure recovery, refer to examples/happy-path.md、examples/boundary-refusal.md、examples/failure-recovery.md。