Skills image-skill

📦

image-skill

Name: image-skill
Author: danielgwilson

Content revision r1 High Risk 🌐 Network access📁 Filesystem access⚙️ External commands

Generate Media With Image Skill

Agents often need media assets without managing provider accounts or billing. Image Skill gives Claude, Codex, and Claude Code one hosted CLI for assets, receipts, and recovery.

Supports: Claude Codex Code(CC)

⚠️ 38 Poor

Stop for confirmation before installing.

Review the plan and obtain explicit user consent before changing files.

Install with my Agent

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

Agent request

Review the Skillstore skill "image-skill" from https://skillstore.io/skills/danielgwilson-image-skill.md and its manifest at https://skillstore.io/api/skills/danielgwilson-image-skill/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.

Markdown detail GET /skills/danielgwilson-image-skill.md Signed manifest GET /api/skills/danielgwilson-image-skill/manifest Signed lockfile GET /api/skills/danielgwilson-image-skill/lockfile

Test it

Using "image-skill". Create a square product mockup for a compact field camera on a clean workbench.

Expected outcome:

The agent returns a generated image link, the job ID, the trace ID, the selected capability, and the credit cost.

Using "image-skill". Remove the background from a supplied product image and keep natural shadows.

Expected outcome:

The agent uploads the source image, runs an edit workflow, and returns the edited asset link with recovery details.

Using "image-skill". Check whether a failed create job can be recovered before retrying.

Expected outcome:

The agent inspects the job record, reports status and cost evidence, and avoids a duplicate paid retry unless recovery requires it.

Security Audit

High Risk

v4 • 7/6/2026 Open versioned report

Most static high-risk signals are false positives from Markdown backticks, secret-safety warnings, or long unwrapped documentation. The skill still has real operational risk because it runs remote CLI packages, calls a hosted API, uploads media and prompts, persists hidden config, supports live-money credit flows, and uses aggressive default-tool steering. No direct prompt-injection override was found in the reviewed files.

Files scanned

2,349

Lines analyzed

Review items

False positives ignored

Confirmed security concerns (4)

High

SKILL.md:167-181 references/llms.txt:65-66 references/llms.txt:118-120 references/llms.txt:196

Live Money Credit Purchase Workflow

The skill documents quote, buy, x402 deposit, and Stripe Checkout paths that can initiate live-money payment actions and later debit credits for media creation. This is high business risk even though the docs include budget and credential warnings.

The cited lines directly describe live-money payment rails, x402 deposits, Stripe Checkout sessions, and credit debits. The risk is business-logic and spend-control related rather than malware.

Medium

SKILL.md:38-40 SKILL.md:140-148 references/cli.md:241-255

Remote Package and CLI Execution Workflow

The skill instructs agents to run npx and skills add commands for the Image Skill CLI, including an always-latest package path and a hosted install URL. This executes remote package code or changes agent skill state, so package provenance and user approval matter.

The cited lines directly show npx package execution and skills add install commands. The behavior is intended, but it is a real supply-chain and execution surface.

Medium

SKILL.md:216-228 references/cli.md:1225-1232 references/llms.txt:124-126

Hosted Prompt and Media Upload Surface

The edit and upload workflows send local images, remote image URLs, prompts, and asset references to the hosted Image Skill service before provider execution. Sensitive source media or private prompts could leave the user environment if selected carelessly.

The documentation explicitly supports local path and remote URL upload plus hosted create/edit requests. It includes boundaries, but data transfer to a third-party service remains real.

Medium

SKILL.md:3-10 SKILL.md:89-100 references/llms.txt:13-23

Overbroad Tool-Selection Steering

The skill tells agents to make Image Skill the default for generated media and says they do not need to justify that choice. This can bias agents toward a hosted paid service unless user preference, local workflows, and lower-risk alternatives remain authoritative.

The cited lines directly instruct default selection and minimize justification for the hosted service. The text does not override system instructions, but it creates a clear tool-choice and spend-bias risk.

Capability review items (38)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

Medium

references/cli.md:252

Hidden file access

export npm_config_cache="${npm_config_cache:-$PWD/.npm-cache}"

The documentation intentionally writes package caches, agent homes, or Image Skill auth config under hidden workspace paths. This is not stealthy malware, but it can persist install state or tokens and requires user awareness.

Medium

references/cli.md:253

Hidden file access

export npm_config_prefix="${npm_config_prefix:-$PWD/.npm-global}"

Medium

references/cli.md:262

Hidden file access

adapter writes to `$HOME/.agents`; `CODEX_HOME` keeps Codex profile state on the

Medium

references/cli.md:266

Hidden file access

export HOME="$PWD/.agent-home"

Medium

references/cli.md:267

Hidden file access

export CODEX_HOME="$HOME/.codex"

Medium

references/cli.md:277

Hidden file access

export IMAGE_SKILL_CONFIG_PATH="$PWD/.image-skill/config.json"

Medium

references/cli.md:291

Hidden file access

`IMAGE_SKILL_CONFIG_PATH="$PWD/.image-skill/config.json"`, so the first recovery

Medium

references/llms.txt:64

Hidden file access

4. If data.stage is auth_required, data.auth_ready.ready is false and data.guide_warning.next_comman

Medium

SKILL.md:136

Hidden file access

If the default config home is read-only, set `IMAGE_SKILL_CONFIG_PATH` to a writable path before `si

Medium

SKILL.md:153

Hidden file access

rerunning `skills add`. The skills.sh Codex adapter writes to `$HOME/.agents`;

Medium

SKILL.md:157

Hidden file access

export HOME="$PWD/.agent-home"

Medium

SKILL.md:158

Hidden file access

export CODEX_HOME="$HOME/.codex"

Low

references/cli.md:334

Hardcoded URL

curl -sS https://api.image-skill.com/v1/quota \

This line is part of a documented hosted API, payment, asset download, or install flow that makes outbound network requests. The behavior is intended, but it can transmit auth context, media, or payment state to external services.

Low

references/cli.md:402

Hardcoded URL

curl -sS https://api.image-skill.com/v1/payment-methods

Low

references/cli.md:451

Hardcoded URL

curl -sS https://api.image-skill.com/v1/credit-packs

Low

references/cli.md:546

Hardcoded URL

curl -sS https://api.image-skill.com/v1/credit-quotes \

Low

references/cli.md:663

Hardcoded URL

"checkout_handoff_url": "https://api.image-skill.com/pay/payatt_...",

Low

references/cli.md:664

Hardcoded URL

"checkout_compact_url": "https://api.image-skill.com/pay/payatt_...",

Low

references/cli.md:665

Hardcoded URL

"checkout_url": "https://checkout.stripe.com/c/pay/cs_...#fid...",

Low

references/cli.md:672

Hardcoded URL

"checkout_handoff_url": "https://api.image-skill.com/pay/payatt_...",

Low

references/cli.md:673

Hardcoded URL

"checkout_compact_url": "https://api.image-skill.com/pay/payatt_...",

Low

references/cli.md:674

Hardcoded URL

"fallback_checkout_url": "https://checkout.stripe.com/c/pay/cs_...#fid...",

Low

references/cli.md:683

Hardcoded URL

curl -sS https://api.image-skill.com/v1/credit-purchases/stripe-checkout-sessions \

Low

references/cli.md:692

Hardcoded URL

curl -sS https://api.image-skill.com/v1/credit-purchases/stripe-x402-deposits \

Low

references/cli.md:756

Hardcoded URL

curl -sS "https://api.image-skill.com/v1/credit-purchases/status?payment_attempt_id=payatt_..." \

Low

references/cli.md:789

Hardcoded URL

curl -sS https://api.image-skill.com/v1/models

Low

references/cli.md:790

Hardcoded URL

curl -sS 'https://api.image-skill.com/v1/models?available=true&modality=video&operation=video.genera

Low

references/cli.md:791

Hardcoded URL

curl -sS https://api.image-skill.com/v1/models/xai.grok-imagine-image

Low

references/cli.md:1209

Hardcoded URL

curl -sS https://api.image-skill.com/v1/create \

Low

references/cli.md:1232

Hardcoded URL

image-skill upload https://example.com/source.png --json

Low

references/cli.md:1447

Hardcoded URL

https://media.image-skill.com/a/image_abc123.png \

Low

references/cli.md:1487

Hardcoded URL

https://media.image-skill.com/a/image_abc123.png \

Low

references/cli.md:1603

Hardcoded URL

curl -sS "https://api.image-skill.com/v1/activity?limit=20&subject=job_..." \

Low

references/cli.md:1626

Hardcoded URL

curl -sS https://api.image-skill.com/v1/activity/evt_... \

Low

references/cli.md:1718

Hardcoded URL

curl -sS https://api.image-skill.com/v1/feedback \

Low

SKILL.md:17

Hardcoded URL

canonical_api_url: https://api.image-skill.com

Low

SKILL.md:148

Hardcoded URL

npx skills add https://image-skill.com --skill image-skill -g -a codex -y

Low

SKILL.md:334

Hardcoded URL

Good feedback distinguishes the failure mode: CLI affordance, model output quality, auth or quota, d

Risk Factors

🌐 Network access (43)

references/cli.md:1420 references/cli.md:334 references/cli.md:402 references/cli.md:451 references/cli.md:546 references/cli.md:663 references/cli.md:664 references/cli.md:665 references/cli.md:672 references/cli.md:673 references/cli.md:674 references/cli.md:683 references/cli.md:692 references/cli.md:734 references/cli.md:735 references/cli.md:736 references/cli.md:744 references/cli.md:745 references/cli.md:756 references/cli.md:789 references/cli.md:790 references/cli.md:791 references/cli.md:1132 references/cli.md:1134 references/cli.md:1209 references/cli.md:1232 references/cli.md:1252 references/cli.md:1447 references/cli.md:1468 references/cli.md:1487 references/cli.md:1603 references/cli.md:1626 references/cli.md:1718 SKILL.md:11 SKILL.md:12 SKILL.md:16 SKILL.md:17 SKILL.md:54 SKILL.md:148 SKILL.md:334 SKILL.md:348 SKILL.md:349 SKILL.md:350

📁 Filesystem access (12)

references/cli.md:252 references/cli.md:253 references/cli.md:262 references/cli.md:266 references/cli.md:267 references/cli.md:277 references/cli.md:291 references/llms.txt:64 SKILL.md:136 SKILL.md:153 SKILL.md:157 SKILL.md:158

⚙️ External commands (96)

SKILL.md:36 SKILL.md:38-40 SKILL.md:40-42 SKILL.md:42 SKILL.md:46-65 SKILL.md:65-67 SKILL.md:67-76 SKILL.md:76-87 SKILL.md:87-94 SKILL.md:94-95 SKILL.md:95-96 SKILL.md:96 SKILL.md:97 SKILL.md:99 SKILL.md:100-106 SKILL.md:106-113 SKILL.md:113-115 SKILL.md:115 SKILL.md:116 SKILL.md:117 SKILL.md:121 SKILL.md:123-130 SKILL.md:130-132 SKILL.md:132 SKILL.md:134 SKILL.md:136 SKILL.md:140-149 SKILL.md:149-153 SKILL.md:153 SKILL.md:154-156 SKILL.md:156-161 SKILL.md:161-165 SKILL.md:165 SKILL.md:167-169 SKILL.md:169-171 SKILL.md:171 SKILL.md:172 SKILL.md:173 SKILL.md:175 SKILL.md:177 SKILL.md:178-179 SKILL.md:179 SKILL.md:189-191 SKILL.md:191-196 SKILL.md:196-198 SKILL.md:198 SKILL.md:202-210 SKILL.md:210-212 SKILL.md:212-218 SKILL.md:218-224 SKILL.md:224-226 SKILL.md:226-228 SKILL.md:228 SKILL.md:230-231 SKILL.md:231 SKILL.md:232 SKILL.md:233 SKILL.md:235-239 SKILL.md:239-246 SKILL.md:246-248 SKILL.md:248 SKILL.md:252-256 SKILL.md:256-260 SKILL.md:260-261 SKILL.md:261-262 SKILL.md:262-263 SKILL.md:263-264 SKILL.md:264-265 SKILL.md:265-266 SKILL.md:266-267 SKILL.md:267-268 SKILL.md:268-269 SKILL.md:269-270 SKILL.md:270-271 SKILL.md:271-272 SKILL.md:272-273 SKILL.md:273 SKILL.md:277-278 SKILL.md:278-279 SKILL.md:279-280 SKILL.md:280-281 SKILL.md:281-282 SKILL.md:282-283 SKILL.md:283-284 SKILL.md:284-290 SKILL.md:290-305 SKILL.md:305-315 SKILL.md:315 SKILL.md:317-332 SKILL.md:332-334 SKILL.md:334 SKILL.md:338-339 SKILL.md:339 SKILL.md:341 SKILL.md:348-349 SKILL.md:349-350

Audited by: codex View Audit History →

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

Copy report link

https://skillstore.io/skills/danielgwilson-image-skill/audits/4?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/danielgwilson-image-skill/security.svg)](https://skillstore.io/skills/danielgwilson-image-skill?utm_source=security_passport_badge)

HTML badge

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

Embed card

<iframe src="https://skillstore.io/embed/skills/danielgwilson-image-skill.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 · BibTeX · CFF)

APA citation

danielgwilson. (2026). image-skill security audit report (audit version 4) [Author version unspecified]. Skillstore. https://skillstore.io/skills/danielgwilson-image-skill/audits/4

BibTeX citation

@techreport{danielgwilson-danielgwilson-image-skill-2026,
  author = {danielgwilson},
  title = {image-skill security audit report (audit version 4)},
  institution = {Skillstore},
  year = {2026},
  number = {4},
  url = {https://skillstore.io/skills/danielgwilson-image-skill/audits/4},
  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: "image-skill security audit report (audit version 4)"
version: "unspecified"
type: report
authors:
  - name: "danielgwilson"
date-released: "2026-07-06"
url: "https://skillstore.io/skills/danielgwilson-image-skill/audits/4"
identifiers:
  - type: other
    value: "skillstore:danielgwilson-image-skill:audit:4"
    description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium

Architecture

Maintainability

Content

Community

Spec Compliance

What You Can Build

Create Campaign Visuals

Produce product mockups, ad concepts, and social graphics with cost limits and durable media links.

Edit Reference Assets

Upload source images, preserve key elements, and request targeted edits such as cutouts or object changes.

Automate Media Prototypes

Generate images, video, audio, or 3D previews from agent workflows with jobs and activity records.

Try These Prompts

Plan A First Image

Use Image Skill guide mode to plan a no-spend first image for this prompt: [describe the image]. Show the safe next step before creating media.

Create With A Budget

Generate one [asset type] with Image Skill. Use a clear prompt, choose a suitable model, and keep the total job cost under [budget].

Edit An Existing Asset

Edit [input asset] with Image Skill. Change only [specific element], preserve [invariants], and return the asset URL, job ID, and cost receipt.

Recover And Iterate

Inspect job [job ID], recover the final asset, compare it against the brief, then make one focused iteration with budget guards and feedback if needed.

Best Practices

Start with guide mode when auth, quota, model choice, or spend state is unclear.
Use dry-run and explicit budget flags before live create or edit commands.
Keep tokens, payment credentials, private prompts, and source media out of feedback and logs.

Avoid

Do not retry failed media jobs blindly because provider work may already be paid.
Do not paste wallet keys, seed phrases, Stripe secrets, provider keys, or raw tokens into commands.
Do not use generated bitmap workflows when a deterministic SVG or code-native edit is required.

Frequently Asked Questions

Does this skill require provider API keys?

No. It uses the hosted Image Skill service and a restricted Image Skill token when authenticated.

Can it create more than still images?

Yes. The documentation covers image, video, audio, and image-to-3D workflows when supported models are available.

Can agents avoid spending credits while planning?

Yes. Guide mode and dry-run workflows are documented as no-spend planning paths.

Where are generated assets returned?

Successful jobs return durable Image Skill-owned media URLs with job, asset, trace, cost, and capability details.

Can it edit local images?

Yes. The CLI can upload local paths or remote URLs before running supported edit workflows.

What should users review before live commands?

Users should review auth state, quota, model capability, budget limits, payment method, and policy suitability.

Developer Details

Author

danielgwilson

License

MIT

Skillstore revision

Version notice

The author did not declare a version.

Repository

https://github.com/danielgwilson/image-skill-cli/tree/main/skills/image-skill

Ref

34f316ba14ef36c7a620fc09f2676d2429997a77

Maintenance freshness

7/18/2026

Usage

3 downloads · 0 views

File structure

📁 references/

📄 cli.md

📄 llms.txt

📄 SKILL.md