arpit arya
Community@arpitarya · India
Fux is a portable knowledge engine for indexing, searching, and governing project documentation with durable rules, ADRs, and graph-based traceability.
Agent Skills by arpit arya
Showing 27 vetted skills indexed across 1 GitHub repositories.
fux-adr
Captures architecture decisions as indexed, code-linked Fux adr entries.
fux-trace
Traverse a knowledge graph to explain how a feature spans modules.
fux-savings
Measures token and dollar cost savings of Fux from real project file sizes.
fux-plan
Converts feature requests into requirements, design, and task entries stored as durable Fux records.
fux-ingest
Batch-ingest URLs, PDFs, spreadsheets, and API specs into a governed draft rule review queue.
fux
Manages project knowledge as frontmatter rules with deterministic index, graph, and memory views.
fux-distill
Converts session decisions into durable Fux memory and ADR entries.
fux-propose-rules
Drafts durable coding rules with rationale from diffs and git history for human ratification.
fux-fetch-rules
Extracts actionable rules from URLs, PDFs, and text files into durable Fux rule entries.
fux-debate
Orchestrates a two-agent adversarial debate to author and ratify constitutional rules.
fux-critic
Critique proposed changes against project principles before they land.
fux-maintain
Keeps a Fux search index current using git hooks, a freshness daemon, and merge drivers.
fux-sources
Manage Fux corpus sources with add, remove, and ingest commands.
fux-config
Diagnose and explain Fux configuration files including fux.toml, tune.toml, and output.toml.
fux-answer
Generates cited answers from a git-committed document index using fux answer and fux verify.
fux-usage
Resolves the fux CLI and routes queries to the correct search verb.
fux-graph
Explore document relationships in a Fux index using explain, graph, and path commands.
fux-inspect
Inspect a Fux search index to report boilerplate, duplicates, unfindable documents, and graph orphans.
fux-index
Bootstrap, diagnose, and rebuild a Fux search index committed to git.
fux-search
Search a committed Fux document index with fux ask, find, and lexical verbs.
fux-correct
Writes human-authored question corrections onto indexed documents using the fux correct command.
fux-enrich
Generates doc2query enrichment questions for documents in a Fux search corpus.
fux-pii
Diagnose and manage PII redaction rules in .fux/pii.toml for the fux search index.
fux-mcp
Serve a Fux search index over MCP with fux_search, fux_passage, and fux_related tools.
Frequently Asked Questions About arpit arya
FAQPage SchemaWhat tasks can I accomplish with arpit arya's Fux skills?▼
You can bootstrap and diagnose a Fux index, search it with ask/find/answer, explore document relationships via graph and path, capture ADRs, distill session decisions into durable memory, propose and debate rules, ingest documents in bulk, measure token-cost savings, and manage PII redaction, fetchers, and decoders.
Who are the Fux skills designed for?▼
They target software engineers and technical leads maintaining documentation-heavy repositories who need cited answers, architecture decision records, spec-driven planning, and governed rule capture. Maintainers benefit from deterministic $0 upkeep via hooks, daemons, and CI index-freshness checks.
How do I install and run Fux in my repository?▼
Run fux setup to bootstrap the index, fux doctor to diagnose it, and fux ingest plus fux build to populate it. The fux-usage skill resolves the fux command across environments, fux-maintain installs hooks and the freshness daemon, and CI can prove the index stays current.
What does Fux cost to run and does it require an LLM?▼
Core maintenance is deterministic and $0: savings measurement, index upkeep, critic passes, and rule governance run without LLM spend. The fux-savings skill audits token and dollar cost wins from your project's real file sizes rather than asserting ROI.
What prerequisites and dependencies do Fux skills need?▼
The fux CLI must be resolvable in your environment; ask and find work offline with no network access, while answer fetches the sources it cites. Optional components include .fux/pii.toml for redaction rules, custom fetchers and decoders written in .fux/, and MCP client registration for editor integration.