Agent
Peer Reviewer 3; evaluates cross-disciplinary relevance, broader impact, and alternative interpretations.
238 tagged knowledge-base, measured the same way as everything else here.
Browse within: home-automation 35privacy 35self-hosted 35fact-checking 31healthcare 31law 31swiss 31knowledge-graph 23documentation 20fastapi 20knowledge-management 19llm-wiki 16claude-code-agents 12claude-code-skills 12
Agent
Peer Reviewer 3; evaluates cross-disciplinary relevance, broader impact, and alternative interpretations.
Agent
Reads a batch of source files and produces FileSummary objects (summary, exports, relationships, tags) for vault-anything's raw layer.
Agent
Writes a category of vault wiki pages as self-contained, Karpathy-style prose with wikilinks.
Agent
Read-only fresh-context reviewer for the merged integration diff across all tasks in an orchestration run. Invoked at Phase 5 so the coordinator's own degraded peak context never has to hold the full integration diff. Returns a fixed VERDICT and FINDINGS.
Agent
Read-only verifier that audits one task's diff and tests for quality. Invoked between self-review and done so the session that wrote the code does not grade its own tests (self-grading guard). Returns a fixed VERDICT and REASONS.
PranavNagrecha/Salesforce-Intelligence
Agent Claude Code
Assesses a batch of Salesforce custom fields for deletability and returns one structured verdict record per field (Keep / Review / Deprecate-then-Remove / Remove) with its evidence. Spawned by /sfi-field-audit, usually several in parallel over batches of 4-8 fields. Use directly when the user asks whether specific…
PranavNagrecha/Salesforce-Intelligence
Agent Claude Code
Adversarially QAs one sf-intelligence component that has already been built and reviewed — runs it, tries to break it, and checks that the disclosures it advertises actually fire. Spawned after sfi-component-reviewer returns SHIP or SHIP-WITH-FIXES, one per component. Use when a component is claimed done, when a…
PranavNagrecha/Salesforce-Intelligence
Agent Claude Code
Reviews a newly built or modified sf-intelligence component — an MCP tool, an extractor, a graph edge producer, a named analysis — and returns a structured verdict on whether it is the right thing, built to this product's grain, and honest about what it does not know. Spawned after a builder finishes and before QA…
Agent
Use proactively whenever work starts in a project directory and Inkwell (mcpinkwellsearch / list / tags) has no hub entry for it. Investigates the project's purpose, stack, and connections to other projects, then writes the required Inkwell hub + linked entries so future sessions (in this project or another) have the…
Agent
The source-intake agent — drains the un-ingested sources queue. Delegate to it when raw material has been dropped into the KB's sources/ directory (scraped pages, whitepapers, transcripts, meeting notes) and needs to be distilled into KB pages, when lint reports unreferenced sources, or on a scheduled intake run.…
Agent
The knowledge-base health agent — the librarian duty. Delegate to it for scheduled KB hygiene runs, lint passes, audits, or whenever KB quality is in doubt (contradictions, stale pages, broken links, unprocessed sources, index drift). Also the right agent after a burst of ingests. Works with personal KBs and…
Agent
The knowledge-base interface agent. Delegate to it for any KB work — answering questions from a markdown knowledge base (queries with citations), writing knowledge into it (ingesting sources, promoting artifacts/reports, filing learnings, correcting pages), or deciding where something belongs in the KB. Works with…
thomascasali/claude-kb-workflow
Agent
Sviluppo backend Laravel (PHP) e Node.js/Express - API, modelli, migrazioni, servizi, queue. Usalo per implementare feature backend nei progetti dell'ecosistema.
thomascasali/claude-kb-workflow
Agent
CI/CD e release - GitHub Actions tag-triggered, build Flutter iOS/Android (runner macOS self-hosted, Codemagic), firma, store deploy, secrets di pipeline. Usalo per pipeline e rilasci.
thomascasali/claude-kb-workflow
Agent
Integrazioni esterne - Stripe, SumUp, email SMTP, push FCM, Telegram, API AI. Usalo per collegare servizi terzi seguendo i pattern KB.
Agent
Parallel batch ingestion agent for the Obsidian wiki vault. Dispatched when multiple sources need to be ingested simultaneously. Processes one source fully (read, extract, file entities and concepts, update index) then reports what was created and updated. Use when the user says "ingest all", "batch ingest", or…
Agent
Comprehensive wiki health check agent. Scans for orphan pages, dead links, stale claims, missing cross-references, frontmatter gaps, and empty sections. Generates a structured lint report. Dispatched when the user says "lint the wiki", "health check", "wiki audit", or "clean up". Context: User says "lint the wiki"…
Agent
How the engineering skills should consume this repo's domain documentation when exploring the codebase.
Agent
Issues and specs for this repo live as markdown files in .scratch/.
Agent
The skills speak in terms of five canonical triage roles. This file maps those roles to the actual label strings used in this repo's issue tracker.
Agent
Knowledge archival role for the ARCHIVE stage — the moat's intake. Extracts durable knowledge from the run's artifacts through the strict extraction gate, classifies each entry, writes the reference-tracking writeback, and produces the archive summary. The only agent that writes knowledge.
Agent
Import step 3 — distill collected sources + profile into at most 13 standardized draft knowledge entries (the strict gate applies), plus an archive summary. The only import agent that writes knowledge.
Agent
Verification role for the BUILDVERIFY, E2EVERIFY and TEST stages. Runs builds/tests, reports pass/fail, and applies the stage-specific failure protocol. Does not change business logic — only verifies. Queries the knowledge base for pitfalls/guidelines(avoid).
Agent
Odin's bulk-ingest worker (ADR-0045). Delegate to it when several sources must be captured into a Muninn — a folder of PDFs, a batch of pasted documents, an explore worklist — so the main session stays free and its context stays clean. Requires the Muninn root and an explicit source list; it will refuse rather than…