The Agent Skills Directory

[COMMAND_EXECUTION]: Uses the Execute tool to run local git commands such as git log and git config to identify code contributors and analyze repository history.
[DATA_EXFILTRATION]: Generates a pre-populated sharing link for X (Twitter), a well-known service, which includes content derived from the analyzed repository. The link is presented to the user and requires manual interaction to be executed.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from the repository being analyzed.
Ingestion points: Repository file system accessed via LS, Grep, Glob, and repository data via Execute (git log).
Boundary markers: Absent; the skill does not use specific delimiters to isolate analyzed code from its own instructions.
Capability inventory: Execute, AskUser, Read, Grep, Glob, LS.
Sanitization: Absent; repository content is processed directly to generate the obituary text.

obituary