The Agent Skills Directory

[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted data from GitHub issue bodies and comments via the gh issue view command. An attacker could embed malicious instructions within an issue to attempt to override the agent's behavior during the resolution process.
Ingestion points: Untrusted external data enters the agent context through the gh issue view output as described in SKILL.md.
Boundary markers: The skill does not define explicit delimiters or instructions to ignore embedded commands within the fetched issue content.
Capability inventory: The agent has the capability to modify local files, execute shell commands via bun, and interact with the GitHub API to comment on or close issues.
Sanitization: There is no evidence of content sanitization or validation of the text retrieved from GitHub issues.
[COMMAND_EXECUTION]: The skill employs the GitHub CLI (gh) for issue management and the bun runtime for executing project-specific tests (bun playwright, bun check). These actions are consistent with the skill's intended purpose of diagnosing and fixing bugs in the author's own repository (udecode/slate).

resolve-slate-issue