Agent skill · Audited from source
Python Flask Web App For Llava Image Captioning Via Replicate Ap
skill · ECNU-ICALK/AutoSkillLicence not declaredAudited 14 Oct 2026 · source 04e912cff54b
Is it safe? With care — CAUTION, grade B (89/100). We read this skill's source at commit 04e912cff54b and found no filesystem, network or shell access and no hard-coded secrets. The audit raised 1 finding, 1 critical or high. What it does: Develop a Python Flask web application that serves an HTML frontend and uses the Replicate API to generate image captions using the LLaVA 13B model.
Cite
Cite this page
ModelsAtlas. “Python Flask Web App For Llava Image Captioning Via Replicate Ap (ECNU-ICALK/AutoSkill)”. Retrieved 14 Oct 2026. https://modelsatlas.com/skills/ecnu-icalk/python-flask-web-app-for-llava-image-captioning-via-replicate-ap
Trust audit
Audit 0.4.1 · source read 14 Oct 2026
CAUTION means the audit found things worth knowing before you trust this skill. Grade B is shared by 16,667 of 17,602 audited skills.
What the source does
- FilesystemNone observedNo reads or writes in the source
- NetworkNone observedNo outbound calls in the source
- ShellNone observedNo commands run in the source
- DependenciesPinnedNothing floats
- SecretsNone foundNo keys in source
Five audit layers
| Layer | What we check | Result | Findings |
|---|---|---|---|
L0 Provenance | Repo owner, licence, archive status, activity | PASS | 0 |
L1 Static source | Secrets, path traversal, supply chain, hidden code | NA | 0 |
L2 Instructions | Prompt injection, authority framing, hidden directives | FAIL | 1 |
L3 Inventory | The files and declarations match the source | PASS | 0 |
L4 Sandbox | Run in isolation and watch behaviour | SKIPPED why | 0 |
Gates applied: no behavioural pass.
Findings (1)
HIGHPrompt injectionprompt.credential_read · SKILL.md:36
7. **Configuration**: Load the Replicate API key from a `.env` file using `python-dotenv`. Set `replicate.api_token` with this key.
- Why
- asks the agent to read credentials
- Fix
- —
Audit 0.4.1 · source sha 04e912cff54b · 14 Oct 2026Report an issue or request a re-scan
Provenance
Who published it, and what the verdict is pinned to
- PublisherECNU-ICALK GitHub organizationOwner of the repository this skill is read from
- Source folderECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/python-flask-web-app-for-llava-image-captioning-via-replicate-apThe exact folder we audited
- LicenceNot declaredAs the repository states it
- Commit pinned for this verdict04e912cff54bA new audit runs when the source changes
- Repository statusNot archivedLast push 157 days ago
- Sandboxed runNot runSkills are read, not executed; layer L4 is skipped
What it tells the agent
Full source on GitHub ↗Its description: Develop a Python Flask web application that serves an HTML frontend and uses the Replicate API to generate image captions using the LLaVA 13B model.
--- id: "8e9821eb-eb46-494d-8fc1-fe7d6c3e8b24" name: "Python Flask Web App for LLaVA Image Captioning via Replicate API" description: "Develop a Python Flask web application that serves an HTML frontend and uses the Replicate API to generate image captions using the LLaVA 13B model." version: "0.1.0" tags: - "python" - "flask" - "replicate-api" - "llava" - "image-captioning" - "web-development" triggers: - "create a flask app for image captioning" - "integrate replicate llava model" - "python web app for image to text" - "use replicate api in flask" --- # Python Flask Web App for LLaVA Image Captioning via Replicate API Develop a Python Flask web application that serves an HTML frontend and uses the Replicate API to generate image captions using the LLaVA 13B model. ## Prompt # Role & Objective You are a Python Web Developer specializing in Flask and AI API integration. Your task is to create a web application that accepts image uploads and returns captions using the Replicate API's LLaVA 13B model. # Operational Rules & Constraints 1. **Backend Framework**: Use Python with Flask. 2. **API Integration**: Use the `replicate` Python library. Do not use raw `requests` to the API endpoints unless explicitly requested. 3. **Model**: Use the `yorickvp/llava-13b` model version. 4. **Input Handling**: Create a POST route (e.g., `/caption`) to accept image files. 5. **Output Handling**: Use `replicate.run` with `stream=True` enabled. Iterate over the output stream to concatenate the full caption. 6. **Frontend**: Serve an `index.html` file using `render_template`. Assume the file is located in a `templates` directory. 7. **Configuration**: Load the Replicate API key from a `.env` file using `python-dotenv`. Set `replicate.api_token` with this key. 8. **CORS**: Enable CORS using `flask_cors`. 9. **Error Handling**: Return appropriate JSON error messages if no file is provided or if the API call fails. # Interaction Workflow 1. Set up the Flask app structure. 2. Implement the `/` route to serve the HTML page. 3. Implement the `/caption` route to handle image processing. 4. Provide the necessary `pip install` commands (Flask, replicate, python-dotenv, flask-cors). ## Triggers - create a flask app for image captioning - integrate replicate llava model - python web app for image to text - use replicate api in flask
Excerpt, first 2,370 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.
Install
Pinned to the audited commit 04e912c
git clone https://github.com/ECNU-ICALK/AutoSkill git -C AutoSkill checkout 04e912cff54b
The author documents no install command, so this uses the repository directly. Our verdict covers commit 04e912cff54b only.
Works in
The author has not said which agents this skill supports, and we have not tested it in any.
Audit history
Every audit, every source change
- !CAUTION B 89 / 100
first audit · source
04e912cff54b
Similar skills
All coding & review skills →ECNU-ICALK/AutoSkill
Executes a 1-back test by comparing the current input to the immediately preceding input in a sequence, reporting the prior item and a match indicator.
ECNU-ICALK/AutoSkill
Identifies the critical 20% of a given topic that yields 80% of the results and creates a focused learning plan to master it.
ECNU-ICALK/AutoSkill
Act as a friendly English teacher for A1 students. Correct errors briefly, ask clarifying questions to maintain dialogue, and strictly limit responses to 100 words.
ECNU-ICALK/AutoSkill
运用理性情绪疗法ABC模型,按客观事实→情绪反应→自动想法→合理/不合理区分四步顺序,引导来访者辨析创伤相关歪曲信念,并通过挑战式提问促发认知重评。
ECNU-ICALK/AutoSkill
Generates academic content components (introductions, conclusions, titles, topics) and full expository essays. Includes specialized capabilities for refining research paper titles with specific constraints, emphasis,…
ECNU-ICALK/AutoSkill
Draft or improve emails to professors using simple vocabulary, friendly tone, and straightforward grammar.
Questions about Python Flask Web App For Llava Image Captioning Via Replicate Ap
What does the Python Flask Web App For Llava Image Captioning Via Replicate Ap skill do?
Is Python Flask Web App For Llava Image Captioning Via Replicate Ap safe to install?
What can Python Flask Web App For Llava Image Captioning Via Replicate Ap access on my machine?
How current is this page?
Sources
- github.com/ECNU-ICALK/AutoSkill/tree/main/SkillBank/ConvSkill/en — the audited source at commit 04e912cff54b and every finding aboveObserved · 14 Oct 2026
Spotted something wrong? Report a correction · Methodology
Keep going in your AI
Opens with this page as the source. We add nothing else to the prompt.
- What does the Python Flask Web App For Llava Image Captioning Via Replicate Ap skill do?
- Is Python Flask Web App For Llava Image Captioning Via Replicate Ap safe to install?
- What can Python Flask Web App For Llava Image Captioning Via Replicate Ap access on my machine?