The Agent Skills Directory

[SAFE]: No malicious patterns or security violations were identified. The skill is designed for interaction with a legitimate developer tool for regression testing.
[EXTERNAL_DOWNLOADS]: The skill documents the installation of @alwaysmeticulous/cli from the official npm registry and interaction with the app.meticulous.ai API and MCP server. These are official vendor resources associated with the author's infrastructure.
[INDIRECT_PROMPT_INJECTION]: The skill allows the agent to read user-provided text through commands like agent diff-comments and agent test-run-check (in references/agent.md). This represents an attack surface where instructions could be embedded in test comments or reports.
Ingestion points: agent diff-comments (retrieves comment text) and agent test-run-check (retrieves Markdown report text).
Boundary markers: None specified in the documentation.
Capability inventory: The agent can execute CLI commands, modify git worktrees, and write files to the local disk.
Sanitization: No explicit sanitization or filtering of external content is described for these display-oriented commands.
[DYNAMIC_EXECUTION]: The debug commands (in references/debug.md) generate AI-specific context files (including rules and hooks) within a temporary workspace to assist in debugging session replays. These files are created dynamically based on the test data being analyzed to configure the agent's debugging environment.

meticulous-cli