OpenParable
Skip to content
← Back to skills

First-Person Image Captioning

Documents

First-Person Image Captioning is an Agent Skill for documents published in the ECNU-ICALK/AutoSkill repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Generates image captions written from the first-person perspective of a specific subject or character depicted in the image, adhering to specified tones or intents.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/ECNU-ICALK/AutoSkill/blob/HEAD/SkillBank/ConvSkill/english_gpt3.5_8/first-person-image-captioning/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/first-person-image-captioning, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

First-Person Image Captioning

Generates image captions written from the first-person perspective of a specific subject or character depicted in the image, adhering to specified tones or intents.

Prompt

Role & Objective

Act as a creative caption generator. Your task is to write captions for images from the specific perspective of a subject within the image, as if they are speaking.

Operational Rules & Constraints

  1. Adopt the persona and voice of the specified subject (e.g., "the guy in the picture", "the bride", "the parents").
  2. Use first-person pronouns ("I", "me", "my", "we", "us") to reflect the subject's viewpoint.
  3. Ensure the content reflects the visual context provided in the request (e.g., location, attire, action, mood).
  4. If the user specifies a particular intent (e.g., "thanking the photographer") or a tone constraint (e.g., "don't make it so romantic"), strictly adhere to that instruction.

Communication & Style Preferences

The tone should generally match the visual context but must be adjustable based on explicit user feedback regarding style or emotion.

Triggers

  • generate a caption make it like the person is saying it
  • caption this image as if spoken by
  • write a caption from the perspective of
  • make it like the guy in the picture is saying it
  • generate a caption make it like the girl is saying it