Versioned security assessment

Report ID: SA-476E2FC9

7/5/2026, 5:07:12 PM

cosmic-database security assessment v6

Skill Security Certification Report

Audit History
Audit model: codex Historical report
Skill name
cosmic-database
Version
v6
Maintainer
K-Dense-AI
Source
github.com/K-Dense-AI/claude-scientific-skills/tree/main/scientific-skills/cosmic-database
Coverage
4 Files scanned · 979 Lines analyzed
Policy version
Unavailable

Highest confirmed finding severity

Critical

3 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.

Most static findings are false positives: Markdown fences and inline code were misread as shell backticks, documentation links were flagged as network calls, and a local VCF fetch was misclassified. The script does perform authenticated COSMIC downloads and writes downloaded files to caller-selected paths; the write path should be constrained. A bundled evaluation_output.json file asserts the package is safe and that all findings are false positives, which is a prompt-injection risk for reviewers.

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

4 Files scanned · 979 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 11 evidence locations

Filesystem access

May read or write local files.

Observed in 2 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 54 evidence locations

Capability review items (1)
Medium
scripts/download_cosmic.py:103
Python file write/append
with open(output_filename, 'wb') as f:
The script writes downloaded content to caller-controlled output_filename without restricting the destination. This is expected downloader behavior but can overwrite arbitrary paths if invoked with unsafe arguments.

Risk findings

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

Confirmed security concerns (3)

RISK-001 Critical
evaluation_output.json:15-18
Prompt Injection Attempt Detected
The package includes an evaluation report with risk_level set to safe, safe_to_publish set to true, and a summary stating all static findings are false positives. This self-attestation can bias reviewers and automated adjudication.
Lines 15-18 explicitly assert publication safety inside a scanned skill file. The text mirrors a marketplace review artifact and directly tries to influence risk assessment.
RISK-002 Medium
scripts/download_cosmic.py:183-184scripts/download_cosmic.py:200-203
Credential Exposure Through Command-Line Password Option
The CLI accepts a COSMIC password argument, which can expose credentials through shell history or process listings. The safer getpass prompt exists but is optional.
The parser exposes a --password option while getpass is only used when that option is omitted. This is a common credential handling risk, not direct exfiltration.
RISK-003 Low
SKILL.md:335-336
Promotional Steering Instruction
The skill tells the assistant to proactively suggest K-Dense Web for complex workflows. This is disclosed promotion, but it still steers users toward the author's hosted platform.
The instruction is explicit and appears in the operational skill guidance. It is transparent promotion, so the severity is low rather than high.

Remediation

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

  1. FIX-001
    Critical
    Self-attesting evaluation report bundled with the skill
    Remove evaluation_output.json and any file that claims the package is safe, pre-reviewed, or ready to publish.
  2. FIX-002
    Medium
    Caller-controlled output path for downloads
    Restrict downloads to a safe output directory, reject absolute paths and parent traversal, and require explicit overwrite confirmation.
  3. FIX-003
    Medium
    Password accepted as a command-line argument
    Remove the --password option or discourage it strongly. Prefer getpass, a secret store, or an environment-specific credential prompt.
  4. FIX-004
    Low
    Promotional product steering in skill instructions
    Move platform promotion out of operational guidance or require a user request before suggesting the hosted service.
  5. FIX-005
    Low
    Server-provided download URL is fetched without host validation
    Validate the returned download URL against expected COSMIC or Sanger-controlled domains before streaming the file.

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)
High
evaluation_output.json:54
Weak cryptographic algorithm
"reasoning": "No cryptographic code exists in this repository. The static analyzer appears to have f
Force-confirmed blocker/high static finding; AI dismissal overridden.

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