Skills ai-automation-workflows
📦
ai-automation-workflows
Content revision r1 Critical ⚙️ External commands🌐 Network access
Automate AI Workflows with inference.sh
AI automation often becomes fragile when scripts mix models, schedules, and retries without structure. This skill provides reusable workflow patterns for batch jobs, pipelines, monitoring, and scheduled generation.
Supports: Claude Codex Code(CC)
Do not auto-install this skill.
The canonical policy requires operator review before any installation action.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Agent request
Review the Skillstore skill "ai-automation-workflows" from https://skillstore.io/skills/inference-sh-9-ai-automation-workflows.md and its manifest at https://skillstore.io/api/skills/inference-sh-9-ai-automation-workflows/manifest. Verify the artifact. Do not auto-install. Inspect the skill and report your findings, then wait for an operator or manual installation decision.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 "ai-automation-workflows". I need weekly social content for five themes.
Expected outcome:
- A weekday content calendar with one theme per day.
- A workflow outline for image and caption generation.
- Operational notes for rate limits, retries, and saved outputs.
Using "ai-automation-workflows". I need to analyze text files every night.
Expected outcome:
- A scheduled data processing plan with input and output folders.
- A model call sequence for summarization and insight extraction.
- Security notes for redaction and external service review.
Using "ai-automation-workflows". I need alerts when an AI job fails.
Expected outcome:
- A monitoring wrapper description with status checks.
- A log strategy with timestamps and runtime duration.
- A webhook alert plan that avoids sending secrets.
Security Audit
Criticalv4 • 7/6/2026 Open versioned report
Most fenced-code, shebang, and formatting matches are false positives caused by Markdown examples. Real risks remain: a pipe-to-shell installer, cron persistence guidance, a dynamic command wrapper, local file disclosure to external AI services, and unredacted webhook alerts.
1
Files scanned
405
Lines analyzed
4
Review items
0
False positives ignored
Confirmed security concerns (5)
Critical
Pipe to shell pattern
curl -fsSL https://cli.inference.sh | sh && infsh login
The quick start fetches a remote install script and pipes it directly to sh. The command executes unaudited remote code before the user can verify it.
High
Cron job modification
crontab -e
The skill instructs users to edit crontab and install recurring jobs. This is legitimate scheduling guidance, but it creates persistence and needs explicit review controls.
High
[HEURISTIC] SUSPICIOUS COMBINATION: Code execution + Persistence mechanism
This combination is common in malware that maintains access
The skill combines executable shell and Python examples with cron-based persistence. The context is automation guidance, but the combination remains security-relevant.
High
Local Files Sent to External AI Service
The data processing template reads text files from ./data/raw and embeds their contents in a prompt sent to an OpenRouter model. This can disclose private data without redaction or consent controls.
The cited lines loop over local text files and include their contents in an external model request. The data sensitivity depends on the user files, but the transfer behavior is explicit.
High
Unredacted Error Data Sent to Webhook
The alert template posts raw command output and command text to a webhook. Errors and arguments often contain paths, tokens, or private request data.
The webhook payload includes raw error output and command arguments. The destination is a placeholder, but the disclosure pattern is concrete.
Capability review items (4)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
High
Shell command substitution
\"prompt\": \"Analyze this data and provide key insights in JSON format: $(cat $file)\"
The template reads local file contents and places them into a prompt sent to an external model. That creates a real data disclosure risk if files contain sensitive information.
High
Hardcoded URL
curl -fsSL https://cli.inference.sh | sh && infsh login
The URL is fetched by curl and piped directly into sh on the same line. A compromised endpoint or transport path would execute arbitrary installer code.
High
Hardcoded URL
curl -X POST "https://your-webhook.com/alert" \
The alert example posts raw error output and command text to an external webhook. That can disclose secrets, paths, or private request data.
Medium
Shell command substitution
result=$("$@" 2>&1)
The wrapper executes whatever argv is passed through "$@" and captures the output. The example passes infsh, but caller-controlled arguments would become local command execution.
Risk Factors
⚙️ External commands (66)
SKILL.md:298 SKILL.md:15-22 SKILL.md:22-24 SKILL.md:24-32 SKILL.md:32-50 SKILL.md:50-56 SKILL.md:56-87 SKILL.md:87-93 SKILL.md:93-110 SKILL.md:110-116 SKILL.md:116-143 SKILL.md:143-149 SKILL.md:149-179 SKILL.md:179-185 SKILL.md:185-197 SKILL.md:197-201 SKILL.md:201-223 SKILL.md:223-229 SKILL.md:229-258 SKILL.md:258-262 SKILL.md:262-283 SKILL.md:283-287 SKILL.md:287-327 SKILL.md:327-333 SKILL.md:333-356 SKILL.md:356-360 SKILL.md:360-376 SKILL.md:376-390 SKILL.md:390-402 SKILL.md:402-404 SKILL.md:20 SKILL.md:64-66 SKILL.md:70-72 SKILL.md:76-78 SKILL.md:82-83 SKILL.md:123-125 SKILL.md:161 SKILL.md:170 SKILL.md:205 SKILL.md:233 SKILL.md:236 SKILL.md:242 SKILL.md:246 SKILL.md:255 SKILL.md:256 SKILL.md:268 SKILL.md:368 SKILL.md:372 SKILL.md:15-22 SKILL.md:56-87 SKILL.md:116-143 SKILL.md:149-179 SKILL.md:201-223 SKILL.md:229-258 SKILL.md:262-283 SKILL.md:360-376 SKILL.md:33 SKILL.md:57 SKILL.md:94 SKILL.md:117 SKILL.md:150 SKILL.md:202 SKILL.md:230 SKILL.md:263 SKILL.md:334 SKILL.md:361
🌐 Network access (5)
Detected Patterns
Pipe to shell patternCron job modification
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.
Copy report link
https://skillstore.io/skills/inference-sh-9-ai-automation-workflows/audits/4?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/inference-sh-9-ai-automation-workflows?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/inference-sh-9-ai-automation-workflows?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/inference-sh-9-ai-automation-workflows/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/inference-sh-9-ai-automation-workflows.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
inference-sh-9. (2026). ai-automation-workflows security audit report (audit version 4) [Author version unspecified]. Skillstore. https://skillstore.io/skills/inference-sh-9-ai-automation-workflows/audits/4BibTeX citation
@techreport{inference-sh-9-inference-sh-9-ai-automation-workflows-2026,
author = {inference-sh-9},
title = {ai-automation-workflows security audit report (audit version 4)},
institution = {Skillstore},
year = {2026},
number = {4},
url = {https://skillstore.io/skills/inference-sh-9-ai-automation-workflows/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: "ai-automation-workflows security audit report (audit version 4)"
version: "unspecified"
type: report
authors:
- name: "inference-sh-9"
date-released: "2026-07-06"
url: "https://skillstore.io/skills/inference-sh-9-ai-automation-workflows/audits/4"
identifiers:
- type: other
value: "skillstore:inference-sh-9-ai-automation-workflows:audit:4"
description: "Skillstore immutable audit report identifier"
Compare variants
2 installable variantsEach author remains a separate installable skill. The recommended variant is ranked by Skillstore evidence.
Why this variant is first
Higher Skillstore usage
Author / source Skillstore Score Evidence Confidence Skillstore usage Updated
inference-sh Recommended
inference-sh-ai-automation-workflows
Skillstore Score 38
Evidence Confidence Medium
Skillstore usage 12
Updated
2026-08-21
inference-sh-9 Current
inference-sh-9-ai-automation-workflows
Skillstore Score 38
Evidence Confidence Medium
Skillstore usage 7
Updated
2026-08-21
Skillstore Score
Why this score Evidence Confidence: Medium 55
Architecture
85
Maintainability
87
Content
69
Community
83
Spec Compliance
What You Can Build
Plan Recurring Content Workflows
Create repeatable pipelines for images, captions, articles, and social posts with clear outputs and retry behavior.
Process Text Data with AI
Design scheduled file processing that summarizes documents, extracts insights, and writes structured results for later review.
Harden AI Batch Jobs
Add logging, alerts, rate limits, fallbacks, and idempotent outputs to scripted model workflows.
Try These Prompts
Begin Batch Automation
Use this skill to create a beginner batch workflow for [items]. Include input format, model choice, rate limit delay, and output naming.
Design a Sequential Pipeline
Use this skill to design a pipeline that turns [topic] into [deliverables]. Include each AI call, dependencies, failure handling, and outputs.
Add Scheduling and Monitoring
Use this skill to convert [workflow] into a scheduled job. Include schedule, logs, alerts, retries, and safe shutdown steps.
Build a Production Workflow Plan
Use this skill to design an advanced AI automation system for [business process]. Include model routing, privacy checks, idempotency, observability, and rollback steps.
Best Practices
- Review every generated script before running it on a workstation or server.
- Redact secrets and private data before sending prompts to hosted AI services.
- Use logs, retries, and idempotent outputs for every scheduled workflow.
Avoid
- Piping remote install scripts directly into a shell without verification.
- Running cron jobs without owners, disable steps, or log rotation.
- Sending raw files, errors, or command arguments to webhooks without redaction.
Frequently Asked Questions
Does this skill run workflows by itself?
No. It provides patterns and prompts. You must configure scripts, credentials, schedules, and execution permissions.
Which tools does it target?
It targets the inference.sh CLI, bash scripts, cron scheduling, webhooks, and a Python subprocess example.
Can I use it with Claude Code or Codex?
Yes. The skill supports Claude, Codex, and Claude Code according to its marketplace metadata.
Does it require an inference.sh account?
The examples assume the inference.sh CLI is installed and authenticated before workflow commands run.
Can it process sensitive data?
It can describe data workflows, but sensitive data needs redaction, approval, and service review before external model calls.
Is it safe to copy the shell examples directly?
No. Review paths, permissions, secrets, install commands, and network destinations before running any example.
Developer Details
Author
inference-sh-9License
MIT
Skillstore revision
r1
Version notice
The author did not declare a version.
Ref
a06681402992ceae98ba04d54cfd4ab004862696
Maintenance freshness
7/18/2026
Usage
5 downloads · 141 views
File structure
📄 SKILL.md