The Agent Skills Directory

[SAFE]: The skill performs expected operations for a knowledge management tool. It uses local scripts to interface with a project's context tree and provides clear, security-conscious instructions for handling credentials and sensitive data through its internal disclosure model.\n- [EXTERNAL_DOWNLOADS]: The documentation references downloading and installing tool updates via npx -y skills add campfirein/skills. These resources are provided by the official vendor 'campfirein' and represent standard installation/update procedures.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external topics which could contain arbitrary content. It mitigates risk by requiring the agent to use structured citation blocks and providing detailed instructions on how to interpret and quote retrieved memory versus following system instructions. Mandatory Evidence Chain: 1. Ingestion points: scripts/query.mjs and scripts/brv.mjs read. 2. Boundary markers: citation_block (pre-formatted by the engine), 📚 From ByteRover: prefix. 3. Capability inventory: scripts/record.mjs (file write), scripts/auth.mjs (network/auth), scripts/sync.mjs (network sync). 4. Sanitization: Structured HTML schema (<bv-topic>) and specific instructions to use the retrieval snippets verbatim.\n- [COMMAND_EXECUTION]: The skill makes extensive use of the node tool to execute bundled JavaScript scripts. These scripts (query.mjs, record.mjs, etc.) are part of the skill's core functionality for managing its data store.

byterover