Versioned security assessment

Report ID: SA-099AFE5D

6/30/2026, 1:01:39 AM

translate security assessment v2

Skill Security Certification Report

Audit History
Audit model: codex Historical report
Skill name
translate
Version
v2
Maintainer
FuHehe12
Source
github.com/FuHehe12/Nitro-Engine/tree/main/translate
Coverage
9 Files scanned · 1,379 Lines analyzed
Policy version
Unavailable

Highest confirmed finding severity

Medium

2 confirmed security findings require attention.

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.

Static analysis reported many high-risk matches, but the reviewed evidence shows most are false positives from Markdown code fences, inline code, examples, and the .md file context. No malware intent, network exfiltration, environment access, or prompt-injection text was found; the confirmed risk is legitimate local filesystem work through Bash commands and Python helper scripts.

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

9 Files scanned · 1,379 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.

Observed in 2 evidence locations

Network access

May connect to external services.

Not recorded by this audit

Filesystem access

May read or write local files.

Observed in 4 evidence locations

Env variables

May read values from the process environment.

Not recorded by this audit

External commands

May invoke commands or programs outside the Skill.

Observed in 4 evidence locations

Capability review items (2)
Medium
SKILL.md:4SKILL.md:130-132SKILL.md:206-208SKILL.md:249-251
Bash Workflow Can Modify Local Files
The skill enables Bash and instructs agents to run local Python utilities and cleanup commands. This is expected for Markdown splitting and merging, but incorrect paths or working directories could alter or remove unintended local files.
The commands are explicit in the workflow and include local cleanup with rm -rf. The risk is operational rather than malicious because paths are intended to target generated translation artifacts.
Medium
scripts/split_md.py:269-343scripts/merge_md.py:216-237
Helper Scripts Write To User-Specified Paths
The split and merge utilities accept input, output directory, manifest, and output file paths, then read or write files through pathlib. This is necessary for the skill, but it creates filesystem impact if a user supplies broad or sensitive paths.
The scripts clearly create directories and write Markdown, context JSON, manifests, and merged output files. No shell injection or network behavior is present, so the risk is bounded to local file writes.

Risk findings

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

Confirmed security concerns (2)

RISK-001 Medium
SKILL.md:158-167references/subagent-prompt.md:40-68
Sub-Agent Translation Prompts Process Untrusted Text
The workflow sends source segments, context files, samples, and tone guidance to sub-agents for translation. No prompt-injection text was found in the skill files, but user-supplied documents could contain instructions that conflict with translation boundaries.
The sub-agent delegation is documented and intentional. The concern depends on future user-provided source content, so confidence is lower than for direct filesystem behavior.
RISK-002 Low
references/user-samples.md:1-55
High Entropy Sample File Is Human-Readable Text
The entropy alert on the sample file is not supported by semantic review. The file contains readable translation examples and style notes, not binary, encrypted, or obfuscated content.
The entire file was readable Markdown with source text, translations, and explanations. No encoded payload, binary blob, or loader pattern was present.

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 (1)
Low
references/refinement-guide.md:30-49references/subagent-prompt.md:76-81references/translation-brief-template.md:18-35scripts/split_md.py:77-87scripts/merge_md.py:117-124SKILL.md:130-133
Static Backtick And Weak Crypto Alerts Are False Positives
The reported Ruby backtick execution matches are Markdown fences or inline-code markers, not Ruby or shell execution. The weak-cryptography and reconnaissance alerts appear to be triggered by Markdown text, examples, file extensions, or ordinary control-flow code, not cryptographic use or system probing.
Reviewed locations show Markdown syntax, documentation, or regex checks for fenced code blocks. No evidence was found for cryptographic APIs, Ruby execution, or reconnaissance commands.

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