The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (gh) for listing, creating, and labeling issues, and provides templates for these commands in SKILL.md and references/github-labels.md.
[EXTERNAL_DOWNLOADS]: The research-agent uses web search and fetch tools to gather data from arbitrary third-party competitor websites, as described in SKILL.md and examples/competitor-profile.md.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it ingests and processes untrusted data from external sources.
Ingestion points: Competitive research fetched via web tools and existing GitHub issue content retrieved during synchronization.
Boundary markers: Absent; there are no instructions to use delimiters or specific ignore-tags for external data.
Capability inventory: The skill can execute local gh CLI commands and perform file write operations to the .pm/ directory.
Sanitization: Absent; there is no evidence of escaping or validating external content before incorporating it into analyses or new GitHub issues.

product-management