Skills controlnet-pose
๐Ÿ“ฆ

controlnet-pose

Content revision r1 High Risk โš™๏ธ External commands๐ŸŒ Network access๐Ÿ“ Filesystem access

Guide Image and Video Generation with Pose References

Text prompts alone can make it hard to reproduce a specific pose or motion. This skill chooses a RunComfy model that uses a control image or reference video to guide generation.

Supports: Claude Codex Code(CC)
Maintain this Skill? Claim it โš ๏ธ 38 Poor

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request
Review the Skillstore skill "controlnet-pose" from https://skillstore.io/skills/prime-skills-controlnet-pose.md and its manifest at https://skillstore.io/api/skills/prime-skills-controlnet-pose/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Test it

Using "controlnet-pose". I have an OpenPose skeleton and want a still illustration in the same stance.

Expected outcome:

Use Z-Image Turbo ControlNet LoRA. Provide a hosted OpenPose control image and a text prompt describing the subject and scene.

Using "controlnet-pose". I want a draft of my illustrated character performing the movement in a rehearsal clip.

Expected outcome:

Use Kling Motion Control Standard for the draft. The rehearsal clip supplies the movement, and the character image supplies the appearance.

Using "controlnet-pose". I need a still image to follow both a pose skeleton and a depth map.

Expected outcome:

Use a multi-condition ComfyUI workflow in RunComfy instead of a single-control CLI route. Prepare the pose skeleton and depth map as separate assets.

Security Audit

High Risk
v1 โ€ข 11/19/2026 Open versioned report

All 62 static alerts are false positives for the behavior they flag, including Markdown backticks, reference URLs, and documented CLI authentication. The skill does recommend unpinned CLI installation, show fragile shell quoting, and omit a confirmation step for cloud disclosure. No evidence of prompt injection or credential theft was found in the reviewed skill.

1
Files scanned
187
Lines analyzed
0
Review items
0
False positives ignored

Confirmed security concerns (3)

High
Unsafe Shell Argument Construction
The Bash examples put JSON inside shell quotes and invite substitution of user prompts and URLs. JSON serialization alone does not protect the surrounding shell argument. Unsafe substitution could result in local command execution.
The examples show the shell quoting boundary, while the security note dismisses it based on CLI behavior. The impact depends on how the agent constructs arguments.
Medium
Unpinned CLI Installation and Execution
Setup suggests installing the latest npm release globally or executing it with automatic confirmation. A compromised release could run with the installing user's permissions.
Both documented commands omit a version, and the alternative uses automatic confirmation. The concern is the installation policy, not evidence of a compromised package.
Medium
Cloud Disclosure Without Explicit Confirmation
The CLI submits prompts and reference URLs to RunComfy. The skill limits inputs to user-provided URLs but does not require approval to share private media or signed links with the service.
The examples identify the data sent, and the security section requires user-provided URLs without a separate disclosure check. No covert destination is shown.
Audited by: codex
Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report
Security Assessment

Copy report link

https://skillstore.io/skills/prime-skills-controlnet-pose/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/prime-skills-controlnet-pose/security.svg)](https://skillstore.io/skills/prime-skills-controlnet-pose?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/prime-skills-controlnet-pose?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/prime-skills-controlnet-pose/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/prime-skills-controlnet-pose.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
Academic citations (APA ยท BibTeX ยท CFF)

APA citation

prime-skills. (2026). controlnet-pose security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/prime-skills-controlnet-pose/audits/1

BibTeX citation

@techreport{prime-skills-prime-skills-controlnet-pose-2026, author = {prime-skills}, title = {controlnet-pose security audit report (audit version 1)}, institution = {Skillstore}, year = {2026}, number = {1}, url = {https://skillstore.io/skills/prime-skills-controlnet-pose/audits/1}, note = {Author version unspecified} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "controlnet-pose security audit report (audit version 1)" version: "unspecified" type: report authors: - name: "prime-skills" date-released: "2026-11-19" url: "https://skillstore.io/skills/prime-skills-controlnet-pose/audits/1" identifiers: - type: other value: "skillstore:prime-skills-controlnet-pose:audit:1" description: "Skillstore immutable audit report identifier"

Compare variants

4 installable variants

Each author remains a separate installable skill. The recommended variant is ranked by Skillstore evidence.

Why this variant is first

Higher Skillstore usage
runcomfy-com Recommended

runcomfy-com-controlnet-pose

Skillstore Score 79
Evidence Confidence High
Skillstore usage 54
Updated

2026-09-09

doany-ai-controlnet-pose

Skillstore Score 79
Evidence Confidence High
Skillstore usage 53
Updated

2026-09-09

agentspace-so-controlnet-pose

Skillstore Score 38
Evidence Confidence High
Skillstore usage 32
Updated

2026-10-22

prime-skills Current

prime-skills-controlnet-pose

Skillstore Score 38
Evidence Confidence Medium
Skillstore usage 1
Updated

2026-11-20

Skillstore Score

Why this score Evidence Confidence: Medium
55
Architecture
85
Maintainability
87
Content
65
Community
91
Spec Compliance

What You Can Build

Illustrate a Specific Stance

Use a prepared OpenPose skeleton to guide a character illustration while the prompt describes the outfit and setting.

Preview Character Motion

Transfer a rehearsal clip onto an approved character image with Kling Standard, then consider Pro for the final version.

Evaluate a Mascot Animation Route

Check whether the community Wan Animate model fits a stylized mascot project that needs audio and pose guidance.

Try These Prompts

Choose a Model for a Still Image
I want to generate a still illustration from an OpenPose skeleton. Recommend a RunComfy model and list the inputs I need. Do not submit a job.
Generate a Pose-Guided Still
Use this approved OpenPose image URL to guide a stylized illustration of a gardener. Confirm the model and data sent to RunComfy before submitting. Put the result in ./out.
Draft a Motion Transfer
Use my rehearsal video for motion and my character image for appearance. Recommend whether Kling Standard is suitable for the draft. Ask before submitting either URL to RunComfy.
Plan a Multi-Control Workflow
I need both an OpenPose skeleton and a depth map to guide one image. Recommend a RunComfy ComfyUI workflow and list the assets to prepare. Do not submit a CLI job.

Best Practices

  • Use a well-framed motion reference with one subject, one continuous action, and no scene cuts.
  • Check the current model schema to confirm that the LoRA supports your control image type.
  • Confirm what will be sent to RunComfy and use only media you have permission to share.

Avoid

  • Passing a raw photo to a route that expects a precomputed pose skeleton.
  • Sending both a pose map and a depth map to a documented single-control CLI route.
  • Pasting untrusted prompts or media URLs into a shell command.

Frequently Asked Questions

What do I need to run a generation job?
You need the RunComfy CLI, network access, a signed-in RunComfy account, and the inputs required by the selected model.
Which Kling tier should I use for motion transfer?
Use Standard for early drafts and iteration. The skill recommends Pro for final video delivery.
Can the skill extract an OpenPose skeleton from my photo?
No. Create the skeleton with a separate OpenPose or DWPose preprocessor, then provide the control image URL.
Can I guide one image with both pose and depth?
Yes, through a suitable ComfyUI workflow. The documented CLI image route is for a single control.
What information does RunComfy receive?
RunComfy receives the job inputs, including prompts and media URLs when the selected model uses them. Confirm before submitting private assets or signed links.
Where does the generated file go?
The CLI downloads results to the specified local output directory. The documented examples use ./out.

Developer Details

License

MIT

Skillstore revision

r1

Version notice

The author did not declare a version.

Ref

45f28967ece55ae5dfc559d3341aa3db06a755d7

Maintenance freshness

11/20/2026

Usage

0 downloads ยท 0 views

File structure

๐Ÿ“„ SKILL.md