سجل التدقيق

Hidden file in home directory

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

Line 179 states runcomfy login writes an API token to ~/.config/runcomfy/token.json. Persisting credentials in a hidden home config file is real secret-storage behavior even with owner-only permissions.

متوسط

SKILL.md:26-28 SKILL.md:65-72 SKILL.md:76-83 SKILL.md:87-94

Ruby/shell backtick execution

`runcomfy run openai/gpt-image-2/edit` through the local RunComfy CLI.

Line 13 states the skill calls the local RunComfy CLI. That is intentional external process execution and should require user consent before sending data.

متوسط

Ruby/shell backtick execution · 4 حالة

```bash

The fenced block includes an npx command that installs a community skill globally. Running npx or global install commands can execute package code and should not be automatic.

متوسط

Ruby/shell backtick execution

1. **RunComfy CLI** — `npm i -g @runcomfy/cli`

Line 45 instructs installing the RunComfy CLI globally with npm. Global npm installs may run package lifecycle scripts and place commands on PATH.

متوسط

Ruby/shell backtick execution

2. **RunComfy account** — `runcomfy login` opens a browser device-code flow.

Line 46 instructs running runcomfy login, which starts authentication and prepares local credential use. This is expected behavior but security-sensitive.

متوسط

Ruby/shell backtick execution

The skill invokes `runcomfy run openai/gpt-image-2/edit` with a JSON body matching the schema. The C

Line 175 confirms the skill invokes runcomfy run and the CLI posts to the RunComfy model API. This is real external command execution with outbound data transfer.

متوسط

Ruby/shell backtick execution

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

Line 179 describes runcomfy login writing an API token to a local config path. The command is legitimate, but it handles persistent credentials.

متوسط

Hardcoded URL

The skill invokes `runcomfy run openai/gpt-image-2/edit` with a JSON body matching the schema. The C

Line 175 states the CLI posts to the RunComfy model API and downloads result URLs. This is intentional outbound network traffic carrying request data.

متوسط

SKILL.md:16 SKILL.md:22 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:171 SKILL.md:175

Hidden file access

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

Line 179 states the RunComfy token is stored under ~/.config. That is expected CLI credential storage, but it is still access to a hidden credential file.

عوامل الخطر

⚙️ الأوامر الخارجية (38)

🌐 الوصول إلى الشبكة (7)

📁 الوصول إلى نظام الملفات (6)

SKILL.md:40 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:179 SKILL.md:179

دقّقه: codex

إصدار التدقيق 7

9 juil. 2026, 07:53

The skill is Markdown-only, and most static backtick, URL, and traversal detections are documentation false positives. Real risks remain because the skill instructs use of the RunComfy CLI, sends prompts and image URLs to a third-party API, and stores an API token in a hidden config path.

1

الملفات التي تم فحصها

184

الأسطر التي تم تحليلها

16

عناصر المراجعة

0

تم تجاهل الإيجابيات الكاذبة

مخاوف أمنية مؤكدة (1)

متوسط

SKILL.md:175-181

Third-Party Prompt And Image URL Processing

The skill states the CLI posts request data to RunComfy and that image URLs are fetched by the RunComfy model server. Users may expose prompts, image URLs, and generated outputs to a third-party service.

Lines 175 and 181 explicitly describe remote POST requests and server-side fetching of user-provided image URLs.

عناصر مراجعة القدرات (12)

هذه قدرات محلية حقيقية قد يُتوقع وجودها لهذه المهارة، لذا فهي تتطلب مراجعة ولكن لا تُحتسب كسلوك خبيث مؤكد.

مرتفع

Hidden file in home directory

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

Line 179 states runcomfy login writes an API token to ~/.config/runcomfy/token.json. Persisting credentials in a hidden home config file is real secret-storage behavior even with owner-only permissions.

متوسط

SKILL.md:26-28 SKILL.md:65-72 SKILL.md:76-83 SKILL.md:87-94

Ruby/shell backtick execution

`runcomfy run openai/gpt-image-2/edit` through the local RunComfy CLI.

Line 13 states the skill calls the local RunComfy CLI. That is intentional external process execution and should require user consent before sending data.

متوسط

Ruby/shell backtick execution · 4 حالة

```bash

The fenced block includes an npx command that installs a community skill globally. Running npx or global install commands can execute package code and should not be automatic.

متوسط

Ruby/shell backtick execution

1. **RunComfy CLI** — `npm i -g @runcomfy/cli`

Line 45 instructs installing the RunComfy CLI globally with npm. Global npm installs may run package lifecycle scripts and place commands on PATH.

متوسط

Ruby/shell backtick execution

2. **RunComfy account** — `runcomfy login` opens a browser device-code flow.

Line 46 instructs running runcomfy login, which starts authentication and prepares local credential use. This is expected behavior but security-sensitive.

متوسط

Ruby/shell backtick execution

The skill invokes `runcomfy run openai/gpt-image-2/edit` with a JSON body matching the schema. The C

Line 175 confirms the skill invokes runcomfy run and the CLI posts to the RunComfy model API. This is real external command execution with outbound data transfer.

متوسط

Ruby/shell backtick execution

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

Line 179 describes runcomfy login writing an API token to a local config path. The command is legitimate, but it handles persistent credentials.

متوسط

Hardcoded URL

The skill invokes `runcomfy run openai/gpt-image-2/edit` with a JSON body matching the schema. The C

Line 175 states the CLI posts to the RunComfy model API and downloads result URLs. This is intentional outbound network traffic carrying request data.

متوسط

SKILL.md:16 SKILL.md:22 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:171 SKILL.md:175

Hidden file access

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

Line 179 states the RunComfy token is stored under ~/.config. That is expected CLI credential storage, but it is still access to a hidden credential file.

عوامل الخطر

⚙️ الأوامر الخارجية (38)

🌐 الوصول إلى الشبكة (7)

📁 الوصول إلى نظام الملفات (6)

SKILL.md:40 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:179 SKILL.md:179

دقّقه: codex

إصدار التدقيق 6

9 juil. 2026, 05:37

Most static findings are false positives caused by markdown backticks, example URLs, and relative links in documentation. The remaining confirmed risks are expected for this skill: prompts and image URLs are processed by RunComfy, and the RunComfy CLI may store an API token in a hidden user config file.

1

الملفات التي تم فحصها

184

الأسطر التي تم تحليلها

7

عناصر المراجعة

0

تم تجاهل الإيجابيات الكاذبة

مخاوف أمنية مؤكدة (1)

متوسط

SKILL.md:175-181

Third-Party Image Processing Disclosure

The skill sends prompts and image URLs to the RunComfy model API for processing, and the service fetches referenced image URLs. This is the core feature, but users must treat it as third-party processing of potentially sensitive visual content.

The documentation explicitly describes the RunComfy API POST, result downloads, and server-side fetching of third-party image URLs.

عناصر مراجعة القدرات (3)

هذه قدرات محلية حقيقية قد يُتوقع وجودها لهذه المهارة، لذا فهي تتطلب مراجعة ولكن لا تُحتسب كسلوك خبيث مؤكد.

مرتفع

Hidden file in home directory

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

The documentation states that runcomfy login stores an API token under ~/.config/runcomfy/token.json. Mode 0600 reduces exposure, but credential storage in a hidden home directory is a real sensitive-file behavior.

متوسط

Hidden file access

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

The referenced hidden config file contains an API token used by the RunComfy CLI. This is legitimate authentication storage, but still security-relevant because it involves local credential material.

منخفض

SKILL.md:16 SKILL.md:22 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:171 SKILL.md:175

Hardcoded URL

The skill invokes `runcomfy run openai/gpt-image-2/edit` with a JSON body matching the schema. The C

The skill documents a real outbound POST to the RunComfy model API and downloads generated outputs from RunComfy domains. This is expected behavior, but it means user image URLs and prompts leave the local environment.

عوامل الخطر

⚙️ الأوامر الخارجية (38)

🌐 الوصول إلى الشبكة (7)

📁 الوصول إلى نظام الملفات (6)

SKILL.md:40 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:179 SKILL.md:179

دقّقه: claude

إصدار التدقيق 5

9 juil. 2026, 03:09

The high-volume static hits are mostly markdown backticks, placeholder URLs, and relative documentation links rather than executable code or path traversal. The real risks are disclosed use of the RunComfy CLI, third-party network processing, generated output downloads, and local token storage.

1

الملفات التي تم فحصها

184

الأسطر التي تم تحليلها

18

عناصر المراجعة

0

تم تجاهل الإيجابيات الكاذبة

مخاوف أمنية مؤكدة (1)

متوسط

SKILL.md:175-182

Third-Party Image Processing Disclosure Required

The skill sends prompts and user-supplied image URLs to RunComfy for processing. Users may upload or reference sensitive visual content, so marketplace listing should clearly disclose third-party processing.

Lines 175-182 explicitly describe API submission, output downloads, URL fetching by the model server, and third-party content risks.

عناصر مراجعة القدرات (14)

هذه قدرات محلية حقيقية قد يُتوقع وجودها لهذه المهارة، لذا فهي تتطلب مراجعة ولكن لا تُحتسب كسلوك خبيث مؤكد.

مرتفع

Hidden file in home directory

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

The skill documents that RunComfy stores an API token under the user home configuration directory. This is expected credential storage, but it is a real filesystem and secret-handling concern.

متوسط

Ruby/shell backtick execution

`runcomfy run openai/gpt-image-2/edit` through the local RunComfy CLI.

The skill openly instructs use of the RunComfy CLI for its intended image-editing workflow. The command is explicit and service-specific, so this is a real external-command dependency but not hidden execution.

متوسط

Ruby/shell backtick execution

1. **RunComfy CLI** — `npm i -g @runcomfy/cli`

This prerequisite asks users to install or authenticate the RunComfy CLI. It is disclosed and expected, but it remains a real external-command and supply-chain dependency for marketplace users.

متوسط

SKILL.md:65-72 SKILL.md:76-83 SKILL.md:87-94

Ruby/shell backtick execution

2. **RunComfy account** — `runcomfy login` opens a browser device-code flow.

This prerequisite asks users to install or authenticate the RunComfy CLI. It is disclosed and expected, but it remains a real external-command and supply-chain dependency for marketplace users.

متوسط

SKILL.md:47

Ruby/shell backtick execution

3. **CI / containers** — set `RUNCOMFY_TOKEN=<token>` instead of `runcomfy login`.

This prerequisite asks users to install or authenticate the RunComfy CLI. It is disclosed and expected, but it remains a real external-command and supply-chain dependency for marketplace users.

متوسط

Ruby/shell backtick execution · 3 حالة

```bash

The skill openly instructs use of the RunComfy CLI for its intended image-editing workflow. The command is explicit and service-specific, so this is a real external-command dependency but not hidden execution.

متوسط

Ruby/shell backtick execution

The skill invokes `runcomfy run openai/gpt-image-2/edit` with a JSON body matching the schema. The C

The skill openly instructs use of the RunComfy CLI for its intended image-editing workflow. The command is explicit and service-specific, so this is a real external-command dependency but not hidden execution.

متوسط

Ruby/shell backtick execution

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

The skill openly instructs use of the RunComfy CLI for its intended image-editing workflow. The command is explicit and service-specific, so this is a real external-command dependency but not hidden execution.

متوسط

SKILL.md:180-182

Ruby/shell backtick execution

- **Input boundary**: the user prompt is passed as a JSON string to the CLI via `--input`. The CLI d

The skill openly instructs use of the RunComfy CLI for its intended image-editing workflow. The command is explicit and service-specific, so this is a real external-command dependency but not hidden execution.

متوسط

SKILL.md:182

Ruby/shell backtick execution

- **Outbound endpoints**: only `model-api.runcomfy.net` (request submission) and `*.runcomfy.net` /

The skill openly instructs use of the RunComfy CLI for its intended image-editing workflow. The command is explicit and service-specific, so this is a real external-command dependency but not hidden execution.

متوسط

Hidden file access

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

The skill documents that RunComfy stores an API token under the user home configuration directory. This is expected credential storage, but it is a real filesystem and secret-handling concern.

منخفض

SKILL.md:16 SKILL.md:22 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:171 SKILL.md:175

Hardcoded URL

The skill invokes `runcomfy run openai/gpt-image-2/edit` with a JSON body matching the schema. The C

The skill states that the CLI posts requests to RunComfy and downloads generated outputs from RunComfy domains. This is an intended but real third-party network dependency.

عوامل الخطر

⚙️ الأوامر الخارجية (38)

🌐 الوصول إلى الشبكة (7)

📁 الوصول إلى نظام الملفات (6)

SKILL.md:40 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:179 SKILL.md:179

دقّقه: claude

إصدار التدقيق 4

7 juil. 2026, 23:38

No prompt injection or malicious hidden behavior was found in SKILL.md. Most static findings are Markdown formatting, placeholder URLs, documentation links, or disclosed token storage. Confirmed risks are the intended RunComfy CLI execution path and the disclosed outbound RunComfy model API dependency.

1

الملفات التي تم فحصها

184

الأسطر التي تم تحليلها

13

عناصر المراجعة

0

تم تجاهل الإيجابيات الكاذبة

عناصر مراجعة القدرات (10)

هذه قدرات محلية حقيقية قد يُتوقع وجودها لهذه المهارة، لذا فهي تتطلب مراجعة ولكن لا تُحتسب كسلوك خبيث مؤكد.

متوسط

SKILL.md:26-28 SKILL.md:65-72 SKILL.md:76-83 SKILL.md:87-94

Ruby/shell backtick execution

`runcomfy run openai/gpt-image-2/edit` through the local RunComfy CLI.

Line 13 states that the skill calls the local RunComfy CLI. This is an intended external command path that can send user-provided edit inputs to a service.

متوسط

Ruby/shell backtick execution · 4 حالة

```bash

Lines 26-28 document an npx skills add command with global installation. This is a real external command risk because it can execute package installation behavior on the user machine.

متوسط

Ruby/shell backtick execution

1. **RunComfy CLI** — `npm i -g @runcomfy/cli`

Lines 45-46 instruct users to install the RunComfy CLI globally and run runcomfy login. These are external commands with local machine and account effects.

متوسط

Ruby/shell backtick execution

2. **RunComfy account** — `runcomfy login` opens a browser device-code flow.

Line 46 directs users to run runcomfy login, which opens an account authorization flow. This is a real external command and authentication side effect.

متوسط

Ruby/shell backtick execution

The skill invokes `runcomfy run openai/gpt-image-2/edit` with a JSON body matching the schema. The C

Line 175 explains that the skill invokes runcomfy run and the CLI posts requests to the model API. This confirms the skill depends on external command execution and remote processing.

متوسط

Ruby/shell backtick execution

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

Line 179 references runcomfy login as part of credential setup. The command is legitimate and disclosed, but it still changes local authentication state.

منخفض

SKILL.md:16 SKILL.md:22 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:171 SKILL.md:175

Hardcoded URL

The skill invokes `runcomfy run openai/gpt-image-2/edit` with a JSON body matching the schema. The C

Line 175 names the RunComfy model API endpoint used for request submission and generated output download. The endpoint is disclosed and expected, but it is a real outbound network dependency.

عوامل الخطر

⚙️ الأوامر الخارجية (38)

🌐 الوصول إلى الشبكة (7)

📁 الوصول إلى نظام الملفات (6)

SKILL.md:40 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:179 SKILL.md:179

دقّقه: codex

إصدار التدقيق 3

5 juil. 2026, 02:32

Most static findings are false positives from Markdown backticks, code fences, placeholder URLs, and relative links. Confirmed risks are the intended use of external RunComfy CLI commands and third-party processing of prompts and image URLs.

1

الملفات التي تم فحصها

184

الأسطر التي تم تحليلها

13

عناصر المراجعة

0

تم تجاهل الإيجابيات الكاذبة

مخاوف أمنية مؤكدة (1)

متوسط

SKILL.md:175 SKILL.md:181

Third-Party Image and Prompt Processing

The skill states that requests are posted to RunComfy model infrastructure and that image URLs are fetched by RunComfy servers. This is disclosed behavior, but it creates a privacy risk for confidential images, prompts, and asset URLs.

SKILL.md explicitly describes posting to model-api.runcomfy.net and server-side fetching of user-provided URLs. The risk is privacy exposure, not hidden exfiltration.

عناصر مراجعة القدرات (9)

هذه قدرات محلية حقيقية قد يُتوقع وجودها لهذه المهارة، لذا فهي تتطلب مراجعة ولكن لا تُحتسب كسلوك خبيث مؤكد.

متوسط

SKILL.md:26-28 SKILL.md:65-72 SKILL.md:76-83 SKILL.md:87-94

Ruby/shell backtick execution

`runcomfy run openai/gpt-image-2/edit` through the local RunComfy CLI.

This location instructs use of external CLI tooling, so following the skill runs local commands. The command target is fixed, but prompts, image URLs, and output paths still enter the command workflow.

متوسط

Ruby/shell backtick execution · 4 حالة

```bash

This location shows an installation or authentication command for external tooling. Following it can install third-party CLI code or create credentials, so it is a real command-execution concern.

متوسط

Ruby/shell backtick execution

1. **RunComfy CLI** — `npm i -g @runcomfy/cli`

This location shows an installation or authentication command for external tooling. Following it can install third-party CLI code or create credentials, so it is a real command-execution concern.

متوسط

Ruby/shell backtick execution

2. **RunComfy account** — `runcomfy login` opens a browser device-code flow.

This location shows an installation or authentication command for external tooling. Following it can install third-party CLI code or create credentials, so it is a real command-execution concern.

متوسط

Ruby/shell backtick execution

The skill invokes `runcomfy run openai/gpt-image-2/edit` with a JSON body matching the schema. The C

This location instructs use of external CLI tooling, so following the skill runs local commands. The command target is fixed, but prompts, image URLs, and output paths still enter the command workflow.

متوسط

SKILL.md:16 SKILL.md:22 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:171 SKILL.md:175

Hardcoded URL

The skill invokes `runcomfy run openai/gpt-image-2/edit` with a JSON body matching the schema. The C

The prose states the CLI posts requests to model-api.runcomfy.net and downloads results from RunComfy domains. This intentional network behavior can expose sensitive prompts or image URLs to a third party.

عوامل الخطر

⚙️ الأوامر الخارجية (38)

🌐 الوصول إلى الشبكة (7)

📁 الوصول إلى نظام الملفات (6)

SKILL.md:40 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:179 SKILL.md:179

دقّقه: codex

إصدار التدقيق 2

28 juin 2026, 05:52

Static analysis flagged many command, network, filesystem, and blocker patterns in SKILL.md. Review found no prompt injection, malicious intent, weak cryptography, or local path traversal, but the skill intentionally runs a third-party CLI, sends prompts and image URLs to RunComfy, and documents token storage.

1

الملفات التي تم فحصها

184

الأسطر التي تم تحليلها

7

عناصر المراجعة

2

تم تجاهل الإيجابيات الكاذبة

عناصر مراجعة القدرات (4)

هذه قدرات محلية حقيقية قد يُتوقع وجودها لهذه المهارة، لذا فهي تتطلب مراجعة ولكن لا تُحتسب كسلوك خبيث مؤكد.

متوسط

SKILL.md:45-47 SKILL.md:65-71 SKILL.md:175

Third-Party CLI Execution and Remote Image Processing

The skill instructs users to install and run the RunComfy CLI, then submit prompts and image URLs to the RunComfy model API. This is the expected function, but it creates supply-chain, privacy, and external service trust requirements.

The documented command execution and outbound API flow are explicit. Confidence is high for elevated risk, while the same context shows this is functional behavior rather than confirmed malicious activity.

متوسط

SKILL.md:56 SKILL.md:181

Image URLs and Prompts Leave the Local Environment

The skill requires publicly fetchable HTTPS image URLs and states that the RunComfy model server fetches image content. Users editing confidential material could expose prompts, image URLs, or generated outputs to a third-party processor.

The data boundary is directly documented in the skill. The risk depends on user data sensitivity, so it is medium rather than high.

متوسط

SKILL.md:5-13 SKILL.md:26-28 SKILL.md:98-106

RunComfy Token Storage Requires Secret Handling

The skill documents that runcomfy login writes an API token to a hidden file in the user configuration directory, or uses RUNCOMFY_TOKEN in CI. The file mode is documented as owner-only, but token handling still requires normal secret hygiene.

The token storage path and environment variable are explicitly documented. The stated file mode reduces severity, and no evidence of exfiltration was found.

منخفض

Static Command Detections Are Mostly Documentation Syntax

Many external command detections come from markdown backticks, shell examples, inline endpoint names, and prompt examples. They do not show executable code inside the skill file, but the RunComfy CLI commands remain intentional operational behavior.

Line-numbered review shows markdown prose and code examples, not a script that executes automatically. The finding is not fully dismissed because users are instructed to run the CLI.

تم تجاهل الإيجابيات الكاذبة الثابتة (2)

تم تجاهل هذه المطابقات الثابتة بواسطة المراجعة الدلالية أو لأنها طابقت رموزًا خاصة بالمخطط فقط، لذا تُعرض للشفافية لكنها لا تؤثر في درجة الجودة.

منخفض

SKILL.md:40 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:160

Path Traversal and Weak Crypto Alerts Are False Positives

The ../ sequence is a markdown link to a sibling skill, placeholder HTTPS URLs are examples, and the exit code section is not cryptographic code. No evidence found for local traversal logic or weak cryptographic algorithm use.

The reviewed lines contain documentation links, placeholder URLs, and a heading. There is no code path that normalizes, reads, writes, or decrypts data using those strings.

منخفض

SKILL.md:1-184

No Prompt Injection Attempt Found

Review of SKILL.md found no text claiming to override system instructions, skip security analysis, declare pre-approval, or change evaluator behavior.

The complete skill file was reviewed with line numbers. Confidence is high, although this is a human semantic review of one markdown file.

عوامل الخطر

⚙️ الأوامر الخارجية (7)

SKILL.md:13 SKILL.md:26-28 SKILL.md:45-47 SKILL.md:65-71 SKILL.md:76-82 SKILL.md:87-93 SKILL.md:175

🌐 الوصول إلى الشبكة (9)

SKILL.md:16 SKILL.md:22 SKILL.md:56 SKILL.md:69 SKILL.md:80 SKILL.md:91 SKILL.md:171 SKILL.md:175 SKILL.md:181-182

📁 الوصول إلى نظام الملفات (5)

SKILL.md:40 SKILL.md:71 SKILL.md:82 SKILL.md:93 SKILL.md:179

الأنماط المكتشفة

External CLI InvocationRemote URL Fetch and Output Download

دقّقه: codex

إصدار التدقيق 1

6 mai 2026, 08:59

Static analysis detected 79 potential security patterns (external_commands, network, filesystem). After semantic evaluation, all findings are FALSE_POSITIVES. The skill is a CLI wrapper for a legitimate image editing API service. External commands are hardcoded CLI invocations, network activity targets known RunComfy endpoints, and filesystem references are user-provided output directories. No prompt injection detected. Security documentation is comprehensive.

1

الملفات التي تم فحصها

184

الأسطر التي تم تحليلها

4

عناصر المراجعة

1

تم تجاهل الإيجابيات الكاذبة

مخاوف أمنية مؤكدة (4)

متوسط

SKILL.md:66-72 SKILL.md:76-83 SKILL.md:87-94

CLI Command Invocation Pattern

The skill invokes the RunComfy CLI (`runcomfy run openai/gpt-image-2/edit`) which executes external commands. This is by design for a CLI wrapper skill.

CLI commands are hardcoded in documentation, not derived from user input. The skill documents legitimate API access patterns with no injection vectors.

متوسط