The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's workflow involves processing potentially untrusted data from git logs and chat history, which presents a surface for indirect prompt injection.
Ingestion points: Git commit messages and conversation history are used to determine completion status.
Boundary markers: There are no explicit delimiters or markers defined to isolate external data from the agent's core instructions.
Capability inventory: The skill can modify the local repository state via filesystem writes using git add . and git commit.
Sanitization: The instructions do not define sanitization or validation routines for data ingested from the environment.
[COMMAND_EXECUTION]: The skill performs standard shell operations to interact with the git version control system.
Evidence: Instructions utilize git log and git diff for analysis, and git add ., git status, and git commit for persisting updates.
[EXTERNAL_DOWNLOADS]: The skill metadata contains a reference to an external GitHub repository for attribution purposes.
Evidence: The header mentions https://github.com/vfarcic/dot-ai. This is a reference to a well-known service used for sourcing information.

prd-update-progress