Versioned security assessment

Report ID: SA-68B13025

6/30/2026, 3:43:00 AM

agent-tools security assessment v2

Skill Security Certification Report

Audit History
Audit model: codex Historical report
Skill name
agent-tools
Version
v2
Maintainer
inference-sh-7
Source
github.com/inference-sh-7/skills/tree/main/skills/agent-tools/
Coverage
5 Files scanned · 554 Lines analyzed
Policy version
Unavailable

Confirmed finding summary

No confirmed security findings

The completed audit recorded no confirmed security findings. This is not proof that the Skill has no side effects.

Installation context

Historical evidence

This report may not describe the currently installable artifact. Open the current Skill page for install guidance.

Open current Skill page

This report does not block or authorize the manifest or ZIP.

The static analyzer reported many command, network, credential, and filesystem patterns, but most command hits are markdown examples for the documented infsh CLI. The confirmed concerns are pipe-to-shell installation, cloud transmission of user inputs, and API key handling. No evidence found of prompt injection, hidden exfiltration logic, or confirmed malicious intent.

Report position

Historical report

Open audit history before using this report to install.

Audit attestation

Not attestable

The required immutable binding is incomplete.

Human verification

Not verified

No human verification is recorded for this report.

Coverage

5 Files scanned · 554 Lines analyzed

4 items shown for review

Limitations

This report does not claim runtime or sandbox execution and does not prove the absence of side effects.

Evidence chain

Follow the evidence from source binding to the install contract. Available evidence supports verification; it is not a safety guarantee.

  1. Source

    Binding unavailable

  2. Artifact

    Identity incomplete

  3. Audit

    Complete

  4. Install contract

    Open manifest to verify

    Open manifest

Capabilities observed

Observed means this report recorded supporting evidence. Not recorded does not prove that a capability is absent.

Contains scripts

May execute code included with the Skill.

Not recorded by this audit

Network access

May connect to external services.

Observed in 30 evidence locations

Filesystem access

May read or write local files.

Observed in 2 evidence locations

Env variables

May read values from the process environment.

Observed in 3 evidence locations

External commands

May invoke commands or programs outside the Skill.

Observed in 136 evidence locations

Capability review items (4)
Medium
SKILL.md:15-20references/authentication.md:5-6references/authentication.md:43-44references/cli-reference.md:5-6
Remote Installer Piped to Shell
The documentation recommends piping a remote installer from https://cli.inference.sh directly to sh. This can execute unexpected code if the endpoint, network path, or installer supply chain is compromised. A manual installation path is documented, which reduces but does not remove this risk.
The pipe-to-shell command is explicitly documented in multiple files. The risk is clear, although the surrounding text presents it as an installer rather than malicious code.
Medium
SKILL.md:36-53references/running-apps.md:5-18references/running-apps.md:119-129
Cloud AI Runs Can Transmit User Data
The skill is designed to run hosted image, video, LLM, search, 3D, and Twitter apps. Prompts, input files, task data, and generated output URLs may be sent to inference.sh or provider apps, so users need data handling warnings.
The documented commands submit prompts and input files to cloud apps and return hosted output URLs. This is core functionality, but it creates a real privacy and data governance concern.
Medium
references/authentication.md:25-33references/cli-reference.md:68-72
API Key Handling Requires Secret Hygiene
The documentation instructs users to set INFSH_API_KEY for automation and notes that it overrides the config file. Environment variables are common, but API keys can leak through logs, shell history, process listings, or CI settings.
The API key variable is directly documented and is a legitimate authentication pattern. The concern is operational exposure rather than evidence of credential theft.
Low
references/app-discovery.md:5-7references/cli-reference.md:23-46references/running-apps.md:25-38SKILL.md:58-67
Markdown Command Examples Trigger External Command Detections
Most external command findings are backticked markdown examples for the allowed infsh CLI or related skill installation commands. They are user-visible documentation, not hidden executable code inside the skill.
The reviewed locations are markdown command examples and command tables. No hidden script file or automatic execution path was found in these locations.

Risk findings

Confirmed security concerns are separated from items that still need review.

No confirmed security findings were recorded for this completed audit.

Expert evidence

Immutable subject identity, scanner metadata, dismissed matches, and source-level evidence.

Artifact subject

Marketplace commit
Unavailable
Content hash
Unavailable
Tree hash
Unavailable
Skill path
Unavailable
Audit payload hash
Unavailable

Analysis metadata

Audit model: codex

Analysis state: Complete

Scope is limited to the recorded files, lines, methods, and evidence. No runtime or sandbox execution is claimed.

Static false positives ignored (2)
Low
SKILL.md:20-28
Weak Cryptography Detections Appear to Be Checksum References
The reviewed cryptography-related text describes SHA-256 checksum verification for a downloaded binary. SHA-256 is appropriate for checksum verification, and no custom weak cryptographic implementation was found.
The only verified cryptographic context is SHA-256 checksum verification. This strongly indicates the weak-crypto static findings are false positives.
Low
SKILL.md:120-126references/app-discovery.md:110-112references/running-apps.md:146-148
Hardcoded URLs Are Documentation and Service Endpoints
The hardcoded URLs point to inference.sh documentation, installer, distribution, and example output locations. These are expected for a skill that documents an external cloud service.
The URLs are visible documentation links and service examples. They do not show covert data exfiltration or suspicious third-party endpoints.

Verify and export

JSON report CITATION.cff SARIF 2.1.0 Signed installation manifest

The manifest and lockfile bind install artifacts to cryptographic hashes. This integrity claim is separate from the security assessment.

Audit attestation: not_attestable