The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to inspect and evaluate external projects, which involves reading untrusted files, commit history, and command outputs. This creates a vulnerability surface for indirect prompt injection where malicious instructions within the analyzed repository could attempt to subvert the agent's behavior during the assessment.
Ingestion points: Processes fast-track-ai/INDEX.md and various files/artifacts from the target repository during the inspection phase.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for the data being ingested.
Capability inventory: The skill has the ability to read files, write reports to the filesystem, and execute a specific local shell script.
Sanitization: There is no mention of sanitizing or validating the content retrieved from the target project repository.
[COMMAND_EXECUTION]: The skill executes a local shell script at ../ft-init/scripts/fast-track-ai-index.sh with a <target> argument. While this is a local tool invocation, the <target> parameter could serve as an injection vector if the agent does not properly escape the input provided to the shell.

ft-assess