The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's preamble execution block uses a bash command to run check-update.sh from a parent directory relative to the skill's location. This is a standard maintenance pattern for local skill repositories to ensure the tool is up to date.
[DYNAMIC_EXECUTION]: The instructions for the subagent in Step 2 suggest a search strategy that involves locating a script and calling it with python3 (referred to as "AnySearch"). This pattern encourages the agent to execute code found within the environment to perform specialized tasks.
[INDIRECT_PROMPT_INJECTION]: The skill fetches competitor intelligence from external web sources using search tools. This data is processed and summarized into reports, which represents a surface for potential indirect prompt injection from untrusted external content.
Ingestion points: External data enters the context via the WebSearch and mcp__exa__web_search_exa tools during the intelligence gathering phase.
Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore instructions that might be embedded in the fetched search results.
Capability inventory: The skill has access to Bash (for preamble commands), Write (for generating reports in docs/), and Task (for orchestrating subagents).
Sanitization: There are no documented steps for sanitizing or validating the content retrieved from the web before it is incorporated into the agent's summary or the final report.

pm-competitor