Audit History

This is an official Anthropic documentation skill teaching the .claude/plugin-name.local.md configuration pattern. All 224 static findings were adjudicated as false positives: they are Markdown code fences, standard shebangs, read-only YAML frontmatter parsing with sed/grep/awk/jq, and stderr diagnostic echoes. The two high-severity hits (.env/secret and /sys access) are defensive guards inside an example security hook that DENIES access to sensitive paths, not exfiltration. No network access, dynamic code execution, credential handling, or prompt-injection was found; no net-new semantic findings were added.

Static findings were mostly false positives from Markdown code fences, Bash examples, validation messages, and defensive path checks. No explicit prompt injection attempt, credential exfiltration, network access, or malicious command execution was found. One semantic concern remains: examples show local Markdown body text being reused as prompt or control-flow content.

Static findings were mostly false positives from Markdown code fences, Bash examples, validation messages, and defensive path checks. No explicit prompt injection attempt, credential exfiltration, network access, or malicious command execution was found. One semantic concern remains: examples show local Markdown body text being reused as prompt or control-flow content.

Official Anthropic skill reviewed with minimal audit scope. Static analysis flags filesystem and external command patterns because examples read local .claude settings files and use shell parsing, but no malicious behavior or prompt injection was found.

Official Anthropic skill providing documentation on plugin settings pattern. Contains only educational content and utility scripts with safe bash practices including set -euo pipefail, file existence checks, and proper input validation. No network access, no dangerous operations, no external dependencies beyond standard POSIX tools.

Official Anthropic skill providing documentation on plugin settings pattern. Contains only educational content and utility scripts with safe bash practices including set -euo pipefail, file existence checks, and proper input validation. No network access, no dangerous operations, no external dependencies beyond standard POSIX tools.

Official Anthropic skill providing documentation on plugin settings pattern. Contains only educational content and utility scripts with safe bash practices. No network access, no dangerous operations, no external dependencies.

Official Anthropic skill providing documentation on plugin settings pattern. Contains only educational content and utility scripts with safe bash practices. No network access, no dangerous operations, no external dependencies.

Official Anthropic skill providing documentation on plugin settings pattern. Contains only educational content and utility scripts with safe bash practices. No network access, no dangerous operations, no external dependencies.