image-batch-runner
Execute ready image briefs and edits, including product or persona images and batch variants. Use resolved references and return image assets with saved run results.
How do I install this agent skill?
npx skills add https://github.com/postplusai/postplus-skills --skill image-batch-runnerIs this agent skill safe to install?
- Gen Agent Trust Hubpass
The skill is focused on batch image generation using a vendor-specific CLI and includes robust safety logic to handle content policy rejections. It processes user-supplied prompts and image references which creates a surface for indirect prompt injection.
- Socketpass
No alerts
- Snykpass
Risk: LOW · No issues
What does this agent skill do?
Image Batch Runner
Use When
- Persona, concept, or shot inputs already exist and the next step is a hosted image generation or reference-based edit run.
- The output must include local image files plus durable request, response, and manifest records for later QA or video rendering.
Do Not Use When
- The task belongs to ideation, QA, or another released skill listed in the handoff section.
- Required inputs are missing and guessing would change the result.
- Creative classification, model/reference policy, or storyboard logic is still
unresolved. Use
image-generationfirst.
Execution Boundary
- Hosted image generation and edits run through the public
postplus media createverb and are async. A submit records the run handle, current status, and completed artifact metadata (bytes are not auto-downloaded; see the download command below). - This runner validates and executes resolved requests. It must not make creative strategy, task-classification, or reference-policy decisions.
- Unless the user or upstream brief specifies a model, use Nano Banana 2.1:
image-nano-banana-2-1-textfor text-only generation andimage-nano-banana-2-1-editfor reference-bound generation or edits. Preserve an explicit model choice; do not silently switch on failure. - Only edit endpoints accept
--reference-image(default edit endpoint:image-nano-banana-2-1-edit). Text endpoints such asimage-nano-banana-2-1-textreject the flag withUnknown option, so any reference-bound generation must target an edit endpoint, not a text endpoint. - Reference-based edits pass each source image via a repeated
--reference-imageflag. Each value may be a local path, HTTPS URL, existing PostPlus media reference, or data URI. The CLI validates and prepares local media before the single hosted submit; do not pre-upload it or construct a manual request object. - Save a finished image output to disk with
postplus media-file download --reference <output.data.artifacts[0].mediaReference> --output-file <path>. Use the completed result's artifact reference as the download source. - Identifiers and run-local state (
assetId,runId,localAssetDir, manifest paths) are minted or derived by the runner — do not supply them. Read them back from the result for the next handoff.
Source And Path
- Use the requested brief, supplied references and explicit visual constraints.
- Use source files from the active project/client folder first. Do not treat one client directory as the default for all image work.
- Keep internal requests, responses, and manifests under
.postplus; keep final user-facing images and manifests in the active asset folder. If no asset folder exists, choose one explicit workspace path.
Delivery And Handoff
- Deliver the generated images and known cost or cost status. Inspect the finished images only when the user asks for that check.
- If processing is still pending, keep the manifest/request paths and the poll
command
postplus media poll --handle <output.data.id> --output path/to/generation-result.json. Reuse the exact--outputpath from the initial submit. A completed poll atomically replaces the processing JSON at that path with the completed result; prefer the CLI-returned action or resume command for that same operation, honor its wait/recovery boundary, and never submit a replacement job.
Stop Conditions
-
Stop when required user intent, source evidence, or owned input artifacts are missing and guessing would change the result.
-
Isolate definite item-local input errors and typed content-policy rejections; preserve other independent items. Do not soften blocked content to evade a restriction.
-
Shared authentication, balance, configuration or service failures stop new submissions. Continue checking and delivering already submitted items using their original identities. Unknown failure scope stops expansion.
-
Correct definite pre-submission field or format errors without changing the requested intent, mode or output constraints. After submission or an unknown outcome, follow the same operation's structured recovery action. Do not buy a replacement merely because a status query failed.
Public Command Boundary
-
Choose the smallest matching command or workflow from the user input and run it directly.
-
Readiness diagnostics:
postplus doctor --skill image-batch-runner. -
Poll a pending image job:
postplus media poll --handle <output.data.id> --output path/to/generation-result.json. Reuse the initial submit's result path so the completed poll atomically replaces its processing JSON. Prefer the returned resume action and stop when the CLI wait/recovery boundary is reached. -
Use
postplus media schema --jsononly when you need the full endpoint, flag, and enum contract or are repairing an unknown request shape. -
Run the hosted image job with the generated command below; do not use another execution interface.
Batch generation
For one complete requested video in a mode with an explicit duration input, allocate the entire approved script across the fewest supported clips. Prefer its maximum clip duration, with a shorter final clip when needed; respect its minimum duration as well. Set each clip's duration explicitly. Preserve the requested total duration, complete approved dialogue, speaker assignments, pauses, reactions, action order, reference roles and continuity across boundaries.
For source-derived modes, retain their existing schema and media-derived duration.
Independent alternatives are separate requests, not segments to concatenate. Keep each alternative's requested settings, prompt and references.
Submit all authorized independent image or video requests concurrently, then follow their original runs to completion. Do not wait for the first result before submitting the others. Hold a request only for an actual input dependency or an explicit user instruction.
postplus media create image-nano-banana-2-1-text \
--prompt "Describe the result you need" \
--wait \
--output ./result.json
Follow the CLI's structured result and reported next action; do not infer recovery from free-text messages. Use existing PostPlus credits for requested work within existing authorization; estimate-only requests do not execute. Do not recharge, buy subscriptions, publish or change external advertising spend without applicable authorization. An initial reservation is not a final price or a hard cost cap. If a required hard final-cost limit cannot be enforced, do not execute; explain that limit. Correct definite pre-submission field or format errors without changing the requested intent, mode or output constraints. After submission or an unknown outcome, resume the same operation through its returned checkpoint or action; never resubmit uncertain work, repeat exhausted recovery, or switch providers to bypass failure. Keep raw results, operation identifiers and recovery commands in execution records. In ordinary chat, give the result, meaningful limitations, known cost and any necessary user decision; do not narrate uploads, polling, internal identifiers or JSON. When asked how media was made, provide the actual model, submitted prompt in the user language and references actually submitted with their roles.
<!-- END GENERATED EXECUTION EXAMPLE -->- Requested tasks use existing PostPlus credits after the current price and balance are checked. Estimate-only requests do not execute. Stop on insufficient credits; do not recharge, buy a subscription, or repeat an uncertain paid operation.
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/postplusai/postplus-skills/image-batch-runner">View image-batch-runner on skillZs</a>