The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill's static analysis scripts (scripts/run_static_analysis.sh and scripts/run_static_analysis.ps1) download the Forge Static Review Tool (FSRT) binary from the atlassian-labs organization on GitHub. This is an expected operation for the skill's scanning functionality and targets a well-known vendor repository.
[COMMAND_EXECUTION]: The skill executes local shell scripts to automate security scanning. This involves downloading, setting execution permissions, and running external binaries on the host system to analyze Forge project directories.
[PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted third-party source code and manifest.yml files. This creates a surface for Indirect Prompt Injection, where malicious instructions embedded in the analyzed code could attempt to influence the agent's audit report or behavior.
Ingestion points: The skill reads the entire Forge project directory provided in the <forge-project-root-directory> argument.
Boundary markers: While the skill uses structured rules, it does not implement explicit delimiters or 'ignore' instructions when interpolating untrusted code into the analysis context.
Capability inventory: The skill can execute shell commands (scripts/run_static_analysis.sh), read files, and write findings to a dedicated security-audit-artifacts/ directory.
Sanitization: There is no explicit sanitization of the analyzed code before it is processed by the agent's logic.

forge-security-review