Versioned security assessment

Report ID: SA-A8205BA7

6/29/2026, 3:42:24 PM

pubchem-database security assessment v6

Skill Security Certification Report

Audit History
Audit model: codex Historical report
Skill name
pubchem-database
Version
v6
Maintainer
davila7
Source
github.com/davila7/claude-code-templates/tree/main/cli-tool/components/skills/scientific/pubchem-database
Coverage
4 Files scanned · 1,676 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 command execution and weak cryptography patterns, but review found they are false positives from Markdown code fences, inline code, and chemistry terms such as CID, InChI, and SMILES. The confirmed risks are intentional outbound requests to PubChem and optional local file writes for downloaded compound structures, so publication is reasonable with a clear network and filesystem warning.

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 · 1,676 Lines analyzed

3 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.

Observed in 4 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 2 evidence locations

Capability review items (1)
Low
SKILL.md:29-34references/api_reference.md:19-26
Markdown Backtick Findings Are Not Shell Execution
Verdict: FALSE_POSITIVE. The external command detections are Markdown code fences, inline endpoint notation, and example snippets, not Ruby backtick shell execution. No evidence found of subprocess, os.system, eval, or shell command construction in the reviewed files. Confidence: 0.97. Confidence reasoning: The cited lines are documentation formatting or Python examples, and the actual scripts do not invoke a shell.

Risk findings

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

Confirmed security concerns (2)

RISK-001 Medium
scripts/bioactivity_query.py:21-22scripts/bioactivity_query.py:28-49scripts/bioactivity_query.py:55-66scripts/bioactivity_query.py:160-176
Intentional External PubChem API Requests
Verdict: TRUE_POSITIVE. The helper script builds PubChem PUG-REST and PUG-View URLs and sends GET or POST requests through requests. This is expected for a PubChem research skill, but user queries and compound identifiers leave the local environment for NCBI PubChem. Confidence: 0.92. Confidence reasoning: The URLs and request calls are explicit, limited to PubChem endpoints, and match the documented purpose of the skill.
RISK-002 Medium
scripts/compound_search.py:213-236SKILL.md:178-200
Local Structure Downloads Can Overwrite Files
Verdict: TRUE_POSITIVE. The compound download helper accepts a filename and calls pcp.download with overwrite enabled, and the documentation shows writing structure image content to disk. This is legitimate export behavior, but users should choose output paths carefully because an existing file could be replaced. Confidence: 0.86. Confidence reasoning: The write behavior is directly visible and intentional, but it is user-directed rather than covert or malicious.

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
scripts/compound_search.py:57-70scripts/compound_search.py:252-257references/api_reference.md:47-60
Weak Cryptography Findings Are Chemistry Terminology
Verdict: FALSE_POSITIVE. The high severity cryptography matches refer to chemistry identifiers and property names such as CID, InChI, SMILES, and assay IDs. No evidence found of MD5, SHA1, DES, RC4, custom encryption, credential handling, or cryptographic verification. Confidence: 0.94. Confidence reasoning: The surrounding context is chemical data retrieval and molecular representation, not security-sensitive cryptographic code.
Low
SKILL.md:1-6
Prompt Injection Scan Found No Evidence
Verdict: FALSE_POSITIVE for prompt injection concern. Reviewed skill files did not contain override commands, fake system messages, pre-approval claims, or instructions to skip security review. Confidence: 0.90. Confidence reasoning: Targeted search across the files from the static analysis found no suspicious prompt-injection phrases, but absence of evidence cannot prove all future content safe.

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