The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill enables the agent to fetch and analyze content from GitHub issues using cs.analyze_issue and codegraph fetch-issue. This data is provided by external users and may contain malicious instructions.
Ingestion points: Data enters the agent context via GitHub issue bodies, titles, and stack traces as described in SKILL.md and bug-analysis.md.
Boundary markers: The instructions do not specify the use of delimiters or specific guardrail instructions to disregard commands found within issue text.
Capability inventory: The skill provides access to local file systems (codebase indexing), graph database execution (cs.conn.execute), and network connectivity to GitHub via the gh CLI.
Sanitization: There is no explicit mechanism described for sanitizing or escaping the natural language content fetched from external issues before it is processed by the agent.
[COMMAND_EXECUTION]: The skill requires the execution of shell commands for installation and operation, including pip install codegraph-ai, codegraph init, and the gh CLI for GitHub authentication and issue retrieval. These are standard operations for the tool's stated purpose.
[EXTERNAL_DOWNLOADS]: The setup process involves downloading the codegraph-ai package from PyPI and retrieving issue data from GitHub's public API. These operations target well-known services and the vendor's own package repository.

codegraph-qa