The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) as it processes untrusted external data in the form of code diffs and requirement documents.
Ingestion points: The agent ingests data from git diff output and files within docs/specs/ and docs/architecture/ as described in SKILL.md.
Boundary markers: There are no explicit instructions or delimiters defined to separate the data being reviewed from the agent's instructions, nor are there warnings to ignore embedded instructions within the diff content.
Capability inventory: The skill utilizes git (rev-parse, diff, log, worktree) and grep commands in SKILL.md. It lacks network access or persistence capabilities.
Sanitization: No sanitization or filtering of the ingested content is performed before it is passed to the Standards and Spec subagents for analysis.
[COMMAND_EXECUTION]: The skill executes several shell commands to facilitate the code review process.
Evidence: SKILL.md contains instructions to execute git rev-parse, git diff, git log, grep -rl, and git worktree add. While these are standard tools for the skill's intended purpose, they represent the execution of local system commands based on user-supplied or repository-derived arguments (e.g., <base> ref).

code-review