The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides instructions to download and execute an installation script from the official Volcengine GitHub repository to resolve plugin registration errors.
Evidence: bash <(curl -fsSL https://raw.githubusercontent.com/volcengine/OpenViking/main/examples/memory-plugin-shared/install.sh) --harness claude in SKILL.md.
[EXTERNAL_DOWNLOADS]: The skill fetches content from external sources for installation, updates, and documentation.
Evidence: References to raw.githubusercontent.com/volcengine/OpenViking, github.com/volcengine/OpenViking, and docs.openviking.ai throughout SKILL.md.
[COMMAND_EXECUTION]: The skill executes local diagnostic scripts and system commands (such as curl, lsof, and docker logs) to troubleshoot the plugin and server environment.
Evidence: node ${CLAUDE_PLUGIN_ROOT}/scripts/ov-memory-doctor.mjs and various manual debug commands in SKILL.md.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes data from logs and external API responses, which are potential vectors for indirect prompt injection.
Ingestion points: The skill reads ~/.openviking/logs/cc-hooks.log and evaluates responses from /health and /ready endpoints of the OpenViking server.
Boundary markers: There are no explicit instructions to use delimiters or ignore potential commands embedded within the logs or server messages.
Capability inventory: The skill is permitted to use Bash to perform network requests and execute scripts.
Sanitization: No sanitization or filtering of the ingested external content is described before it is processed by the agent.

ov-memory-doctor