Most static findings are false positives from Markdown code fences, tmux syntax, expected dotfile paths, and documented public URLs. The fixed /tmp cache examples and pane-content persistence setting are real local privacy and integrity concerns. No prompt injection or data-exfiltration intent was found.
The plugin setup enables automatic restore and pane-content capture. This can write terminal output, including tokens or commands, into tmux-resurrect state files.
Line 59 directly enables pane content capture, and lines 57-58 enable automatic restore and periodic saves. This is a clear local privacy risk, although it is not exfiltration.
Capability review items (2)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
The example uses a fixed cache file under /tmp, which can be read or replaced by other local users on shared systems. Use a user-owned cache directory with private permissions.
CACHE="/tmp/news-cache" # one headline per line, refreshed by background fetch
The example uses a fixed cache file under /tmp, which can be read or replaced by other local users on shared systems. Use a user-owned cache directory with private permissions.
Most static findings are false positives from Markdown code fences, tmux syntax, expected dotfile paths, and documented public URLs. The fixed /tmp cache examples and pane-content persistence setting are real local privacy and integrity concerns. No prompt injection or data-exfiltration intent was found.
The plugin setup enables automatic restore and pane-content capture. This can write terminal output, including tokens or commands, into tmux-resurrect state files.
Line 59 directly enables pane content capture, and lines 57-58 enable automatic restore and periodic saves. This is a clear local privacy risk, although it is not exfiltration.
Capability review items (2)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
The example uses a fixed cache file under /tmp, which can be read or replaced by other local users on shared systems. Use a user-owned cache directory with private permissions.
CACHE="/tmp/news-cache" # one headline per line, refreshed by background fetch
The example uses a fixed cache file under /tmp, which can be read or replaced by other local users on shared systems. Use a user-owned cache directory with private permissions.
Static analysis found many command, network, and filesystem patterns, but review shows they are mostly tmux setup examples and framework installation guidance. The confirmed risk is medium because following the skill can clone third-party repositories, change files under the user's home directory, create a forced tmux symlink, and configure tmux status command substitutions. No malicious intent or prompt injection attempt was found.
2
Files scanned
227
Lines analyzed
8
Review items
1
False positives ignored
Capability review items (5)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
The skill instructs agents to clone frameworks, copy tmux files, create a forced symlink, install TPM plugins, and source tmux config files. This matches the skill purpose, but it can overwrite or alter a user's existing tmux configuration if applied without review.
The file operations are explicit and semantically tied to tmux setup. The risk is environmental modification, not evidence of malware.
The skill references GitHub clone commands for tmux frameworks and plugins, plus a Yahoo Finance curl request for a widget. These endpoints are relevant to the stated purpose, but they still introduce third-party network trust and availability risks.
The network commands are directly visible and serve documented tmux installation and widget examples. No credential exfiltration or suspicious destination was found.
The skill teaches tmux status entries that execute shell scripts through command substitution. This is a normal tmux feature, but any script path placed there will run repeatedly inside the user's tmux session.
The command-substitution behavior is explicit in tmux configuration examples. It is legitimate but can become unsafe if users insert untrusted scripts.
Markdown Backtick Detections Are Not Ruby Execution
The static analyzer reported many Ruby backtick execution patterns, but the reviewed locations are Markdown code fences or inline command examples. The examples still include shell commands, but the Ruby-specific detection is a false positive.
The files are Markdown documentation, not Ruby source. The backticks delimit examples rather than executing code.
The reported reconnaissance-like patterns are tmux troubleshooting references, such as checking tmux options and known status bar mistakes. They do not enumerate host secrets or collect system identity data.
The commands inspect tmux configuration state for troubleshooting. I found no semantic evidence of host reconnaissance beyond normal tmux diagnostics.
Static false positives ignored (1)
These static matches were dismissed by semantic review or matched schema-only tokens, so they are shown for transparency but do not drive the quality score.
Weak Cryptography Static Findings Are False Positives
The scanner flagged weak cryptography at descriptive text locations, but no cryptographic algorithm or password hashing implementation appears in the reviewed files. No evidence found for real cryptographic behavior.
The cited lines describe the skill and do not contain crypto APIs, hashes, or encryption instructions. This is a strong false positive.
Static analysis detected 125 patterns but all are false positives. Network URLs point to legitimate GitHub repositories for tmux frameworks. External command patterns are standard tmux #() syntax for status bar widgets. Filesystem operations target standard tmux config locations (~/.tmux.conf, ~/.config/tmux/). No malicious intent, credential theft, or data exfiltration detected.