The Agent Skills Directory

[SAFE]: The skill operates entirely on local project assets, such as CSV result files, paper drafts, and internal agent state files (e.g., paper/.agent/result-inventory.md). All file operations are within the expected scope of a research assistant tool.
[SAFE]: No evidence of credential exposure, hardcoded secrets, or unauthorized network operations was found. The skill does not perform any external downloads or remote code execution.
[SAFE]: The use of the Bash and Glob tools is restricted to searching for and listing existing experiment result files within the project's directory structure.
[SAFE]: The instructions do not contain any prompt injection patterns, obfuscation, or attempts to bypass system safety guidelines.

paper-evidence-gap-miner