Versioned security assessment

Report ID: SA-AC1A4684

10/4/2026, 1:37:48 PM

mantis-dedupe security assessment v1

Skill Security Certification Report

Audit History
Scanner version 3.0.0 Audit model: codex Latest published report
Skill name
mantis-dedupe
Version
v1
Maintainer
google
Source
github.com/google/mantis/tree/f48a85e8823ee6f4824ae9a24b9b2efb8aab8c9a/mantis-dedupe
Coverage
1 Files scanned · 372 Lines analyzed
Policy version
skillstore-security-audit-policy-v1

Highest confirmed finding severity

High

2 confirmed security findings require attention.

Installation context

Check the current Skill page

This page summarizes report evidence only. The Skill page provides the canonical install advisory.

Open current Skill page

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

The 160 static alerts describe Markdown notation, legitimate workspace operations, and defensive snapshot checks rather than the alleged execution or reconnaissance threats. Semantic review found unsafe helper reuse and potential disclosure of sensitive vulnerability summaries through external embeddings. No evidence found of malicious intent or instructions to manipulate this audit.

Report position

Latest published report

Latest refers to the report sequence, not to artifact currentness.

Audit attestation

Active attestation

A public attestation is available for this exact report.

Human verification

Not verified

No human verification is recorded for this report.

Coverage

1 Files scanned · 372 Lines analyzed

2 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

    Commit and path bound

  2. Artifact

    Content and tree hashes bound

  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.

Not recorded by this audit

Filesystem access

May read or write local files.

Observed in 16 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 50 evidence locations

Risk findings

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

Confirmed security concerns (2)

RISK-001 High
SKILL.md:312-318SKILL.md:349-350
Version Comment Does Not Authenticate Reused Helpers
The skill says, "Only reuse it when the marker matches." A modified helper can retain the expected comment and execute unauthorized code.
The instructions explicitly authorize helper reuse based on a version comment and later execute it. No content-integrity verification is specified.
RISK-002 Medium
SKILL.md:264-278
External Embeddings May Disclose Vulnerability Details
The semantic fallback embeds root-cause summaries using a configured provider, defaulting to Vertex AI. These summaries include vulnerability mechanisms and dataflow without a disclosure-approval requirement.
The instructions specify sensitive summary fields and an external default embedding provider. Actual transmission depends on configured integration; no evidence found of malicious exfiltration intent.

Remediation

Suggested fixes recorded by this audit. Applying them is the maintainer’s responsibility.

  1. FIX-001
    High
    Existing helper scripts are trusted based only on a reproducible version comment.
    Regenerate helpers from trusted instructions or verify their complete contents against a trusted digest before execution; reject unexpected symlinks and untrusted ownership.
  2. FIX-002
    Medium
    Embedding fallback can send sensitive vulnerability summaries to an external provider without an explicit disclosure gate.
    Require approval before external embeddings, document transmitted fields, redact sensitive details, and provide a local embedding option.

Expert evidence

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

Artifact subject

Marketplace commit
ac1a468421c65e380e46c4c08b5b3b8faaa5aa45
Content hash
4cc3dd33a392ca62bc6685444e6f35bb626daf058c5cc727c3eef258cb5e2d29
Tree hash
9fd9e68309d7d2c4b4d62d8bd8c467a9b2e3252186dfa47af07f749519159617
Skill path
skills/google/mantis-dedupe
Audit payload hash
5e78f5d032f49a554ccc7696d61a3739

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.

Verify and export

JSON report Signed attestation CITATION.cff SARIF 2.1.0 Signed installation manifest Source

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

Audit attestation: active