vp-knowledge

The only one with shears — mends what the gardener marked. Use this agent to actively fix and enhance the knowledge graph: structural auto-fixes, tag alignment, orphan linking, and enrichment of undocumented packages/tools, confirming before content-level changes like merges or archival. Typical triggers include: "fix the issues from the graph audit", "improve my knowledge graph — fix orphans, add missing links, enrich thin notes", a specific structural-fix request (e.g. "a bunch of my npm notes are missing relations, can you fix that?"), or "research and document any important packages that are missing". This is the write-capable counterpart to the read-only knowledge-gardener — invoke this agent, not the gardener, whenever notes must actually change. See "When to invoke" in the agent body for worked scenarios.

Walks the beds before work begins and reports what is already known. Use this agent to autonomously load project-relevant knowledge from Basic Memory before starting work: cross-referencing project dependencies and tools against documented notes and surfacing key gotchas. Typical triggers include: "prime the knowledge graph for this project", "what does Basic Memory know about this project's dependencies", or "load any relevant gotchas before I start working on the auth module". Read-only — it never writes or modifies notes; it is the "before work" counterpart to /session-reflect. See "When to invoke" in the agent body for worked scenarios.

Tends the bookmark annex — same discipline, different soil. Use this agent for read-only Raindrop.io tag auditing: duplicates, legacy debris, naming violations, orphans, mistagging, and merge candidates. Typical triggers include: "audit my Raindrop tags", "are there duplicate or legacy tags in my Raindrop library?", or "which Raindrop tags should I merge or delete?". This agent never modifies tags or bookmarks — it produces recommendations only, for the user to apply. See "When to invoke" in the agent body for worked scenarios.

This skill should be used when the user asks 'what do I know about [topic]', 'what does Basic Memory say about [X]', 'do I have notes on [X]', 'look up [X] in my knowledge graph', 'ask memory about [X]', 'recall what we captured about [topic]', 'find notes on [topic]', 'knowledge question', 'query the knowledge graph', 'what observations do I have about [X]', 'knowledge lookup'. Searches Basic Memory for notes and observations matching a specific topic or question -- synthesizes an answer with source citations and gap suggestions. NOT for project-wide inventory or coverage reports (use /knowledge-prime for that).

This skill should be used when the user asks to 'reflect on this session', 'save insights to memory', 'capture what we learned', 'commit this to memory', 'save decisions to Basic Memory', 'write up what we did', 'preserve what we found', 'save insights from this conversation', 'save what was discovered', 'capture session learnings', 'session reflection'. Reviews the current conversation, extracts durable insights (decisions, lessons, gotchas, patterns), previews proposed captures grouped by target note, and writes to Basic Memory after user approval. This captures durable insights as Basic Memory notes — to save URLs or links found this session as Raindrop bookmarks instead, use session-bookmarks.

This skill should be used when the user asks to 'manage tag vocabulary', 'update raindrop tags', 'sync tag vocabulary', 'curate tags', 'refresh tags file', 'rebuild tag vocabulary', 'what tags should I use', 'tag reference', 'raindrop tag vocabulary', 'create tags reference', 'tag inventory', 'tag sync'. Fetches the user's Raindrop tags, selects the top N by usage count, adds one-line characterizations, groups them by cluster, and writes or syncs the vocabulary file at ~/.claude/references/raindrop-tags.md.

This skill should be used when the user asks about 'knowledge gaps', 'tool coverage', 'undocumented dependencies', 'undocumented tools', 'concept gaps', 'installed plugins', 'plugin coverage', 'undocumented skills', 'globally installed plugins/skills', 'what is installed on this machine', 'stale/outdated/drifted notes', 'version drift', or 'which tools/packages need updating'. Audits project dependency and tool manifests — and installed Claude Code plugins + skills.sh bundles — against Basic Memory coverage, and detects concept-level hub gaps. Two flag modes: `--stale [brew|npm|cask|crate|vscode|plugin]` checks version drift instead of coverage; `--global` audits what is installed on this machine — Claude Code plugins + skills.sh bundles today — against coverage. Supported ecosystems and full flag mechanics are documented in the skill body.

This skill should be used when the user asks to audit, health-check, or structurally validate one or more SPECIFIC named notes or a bounded topic cluster — 'audit these notes', 'check this note for orphans or broken links', 'fourth-wall check on [note]', 'validate the structure of [note]', 'spot-check [note]'. Runs a read-only audit inline. NOT for fixing issues (use /knowledge-maintain), NOT for freeform topic questions (use /knowledge-ask), and NOT for whole-graph audits ('audit my knowledge graph', 'graph health', 'full audit', no arguments) — those belong to the knowledge-gardener agent, which this skill delegates to when invoked graph-wide.

5 hooks across 4 events