The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to search and analyze external codebases and framework documentation. This creates a vulnerability where malicious instructions embedded in the analyzed code or documentation (indirect prompt injection) could influence the agent's behavior.
Ingestion points: The skill ingests raw code chunks, symbol summaries, and documentation from the local environment and external sources via claudemem search and claudemem docs.
Boundary markers: The instructions lack explicit boundary markers or directives for the agent to treat search results as passive data rather than instructions.
Capability inventory: The skill utilizes Bash and Task tools, which provide the ability to execute shell commands and modify the filesystem.
Sanitization: There is no mention of sanitization or safety filtering for the content retrieved during search and mapping operations.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of external software and retrieves documentation from third-party services.
Software installation: The skill instructs the user to install the claude-codemem utility via npm install -g claude-codemem or brew install claudemem.
Documentation providers: The docs fetch command retrieves data from external providers including Context7 (context7.com) and DevDocs (devdocs.io).

claudemem-search