Versioned security assessment

Report ID: SA-830451BC

6/28/2026, 10:19:03 AM

Agent Development security assessment v5

Skill Security Certification Report

Audit History
Audit model: codex Historical report
Skill name
Agent Development
Version
v5
Maintainer
anthropics
Source
github.com/anthropics/claude-plugins-official/tree/main/plugins/plugin-dev/skills/agent-development
Coverage
8 Files scanned · 2,603 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 high-risk patterns, but review found no malicious behavior or prompt injection. Most detections are markdown code fences, placeholder ellipses, security-review examples, or read-only validation commands. The skill is publishable with low risk because it includes a local shell validator that reads a user-provided path.

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

8 Files scanned · 2,603 Lines analyzed

1 item 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 3 evidence locations

Network access

May connect to external services.

Not recorded by this audit

Filesystem access

May read or write local files.

Observed in 3 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 (1)
Low
scripts/validate-agent.sh:1-5scripts/validate-agent.sh:20-29scripts/validate-agent.sh:48-49
Read-Only Shell Validator Requires Normal Script Caution
The skill includes a Bash validator that accepts an agent file path, verifies that it exists, and parses its contents with standard local tools. The path is quoted and the script does not write files, execute file contents, or perform network activity, so this is a low operational concern rather than malicious behavior.
The script behavior is directly visible and limited to local validation. Confidence is high, but users should still review any shell script before running it.

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:24-58examples/agent-creation-prompt.md:19-23scripts/validate-agent.sh:33-49
Static External Command Alerts Are Markdown and Benign Parsing
The analyzer flagged many backticks in markdown files and command substitutions in the validator. The markdown backticks are formatting delimiters for examples, and the Bash substitutions read or parse a quoted local file path without untrusted command construction.
The flagged markdown regions are code fences, not executable Ruby or shell. The script uses fixed commands such as head, sed, awk, and grep with quoted variables.
Low
scripts/validate-agent.sh:49examples/agent-creation-prompt.md:146references/agent-creation-system-prompt.md:87
Sensitive, Crypto, And Traversal Alerts Are Documentation False Positives
The sensitive-file and weak-cryptography alerts appear to be substring matches inside words, markdown paths, or security-review examples. The reviewed files do not contain credential theft, Windows SAM access, cryptographic implementation, or path traversal logic.
The cited lines are parser variables or placeholder example text. No filesystem traversal, SAM database access, or weak cryptographic API usage was found in the reviewed context.

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