arckit-aws-research-readerReader subagent invoked by /arckit:aws-research (orchestrator). Queries the AWS Knowledge MCP for one research category and extracts factual evidence about AWS services, including region availability for the project's required regions. Returns a JSON payload conforming to arckit-claude/schemas/cloud
arckit-aws-researchUse this agent when the user needs AWS-specific technology research using the AWS Knowledge MCP server to match project requirements to AWS services, architecture patterns, Well-Architected guidance, and Security Hub controls. Examples: <example> Context: User has a project with requirements and wan
arckit-azure-research-readerReader subagent invoked by /arckit:azure-research (orchestrator). Queries the Microsoft Learn MCP for one research category and extracts factual evidence about Azure services, including region availability for the project's required regions. Returns a JSON payload conforming to arckit-claude/schemas
arckit-azure-researchUse this agent when the user needs Azure-specific technology research using the Microsoft Learn MCP server to match project requirements to Azure services, architecture patterns, Well-Architected guidance, and Security Benchmark controls. Examples: <example> Context: User has a project with requirem
arckit-cloud-research-writerWriter subagent invoked by the /arckit:aws-research, /arckit:azure-research and /arckit:gcp-research orchestrators. Renders a validated, scored payload into an AWRS, AZRS or GCRS artefact under projects/{P}-{NAME}/research/. Has no web/MCP/Agent tools — can only render structured input it is given.
arckit-competitors-writerWriter subagent invoked by the /arckit:competitors orchestrator command. Renders a validated, orchestrator-prepared payload into a CMPT Competitor Landscape artefact under projects/{P}-{NAME}/research/, and spawns/enriches per-vendor Government Award History profiles. Has no web/MCP/Agent tools. Not
arckit-competitorsUse this agent when the user needs a competitor landscape for a UK public-sector market — rival suppliers, awarded-value market share, head-to-head comparison against a focal supplier, and market concentration — drawn from the UK Tenders MCP. Examples: <example> Context: User wants to know who a ven
arckit-datascout-readerReader subagent invoked by arckit-datascout (orchestrator). Fetches and extracts factual evidence about external data sources for one (category, source_type) pair. Returns a JSON payload conforming to arckit-claude/schemas/datascout-handoff.schema.json. Not user-invocable — only the orchestrator dis
arckit-datascout-writerWriter subagent invoked by arckit-datascout (orchestrator). Renders a validated, scored payload into a DSCT artefact under projects/{P}-{NAME}/research/. Has no web/MCP/Agent tools — can only render structured input it is given. Not user-invocable — only the orchestrator dispatches this subagent via
arckit-datascoutUse this agent when the user needs to discover external data sources — APIs, datasets, open data portals, and commercial data providers — to fulfil project requirements. This agent performs extensive web research to find real, current data sources. Examples: <example> Context: User has a project wit
arckit-evaluate-readerReader subagent invoked by the /arckit:evaluate and /arckit:score orchestrator commands. Reads one supplier's documents and extracts bounded claims against the evaluation criteria. Returns a JSON payload conforming to arckit-claude/schemas/evaluate-handoff.schema.json. Has no network or write tools.
arckit-evaluate-writerWriter subagent invoked by the /arckit:evaluate and /arckit:score orchestrator commands. Saves a supplier's validated claims and renders the orchestrator's scores into evaluation.md or scores.json. Has no web, MCP or Agent tools, and doesn't read supplier documents. Not user-invocable — only the orc
arckit-evaluatePrompt for /arckit:evaluate on runtimes with no subagents. Create vendor evaluation framework and score vendor proposals, reading supplier documents in one context. Not for dispatch in Claude Code: run the /arckit:evaluate command, which keeps bids out of the scoring context.
arckit-frameworkUse this agent when the user wants to transform existing project artifacts into a structured framework with phased organization, an overview document, and an executive guide. This agent reads all project artifacts and synthesises them into a coherent framework structure. Examples: <example> Context:
arckit-fresh-eyes-reviewerFresh-eyes reviewer started by /arckit:hld-review and /arckit:dld-review. Reads a design with no knowledge of the session that started it and reviews it through one lens: comprehension, principles, requirements, implementation or hld. Returns a JSON object of findings. Has no write, network or deleg
arckit-gcp-research-readerReader subagent invoked by /arckit:gcp-research (orchestrator). Queries the Google Developer Knowledge MCP for one research category and extracts factual evidence about Google Cloud services, including region availability for the project's required regions. Returns a JSON payload conforming to arcki
arckit-gcp-researchUse this agent when the user needs Google Cloud-specific technology research using the Google Developer Knowledge MCP server to match project requirements to Google Cloud services, architecture patterns, Architecture Framework guidance, and Security Command Center controls. Examples: <example> Conte
arckit-gov-code-search-readerReader subagent invoked by /arckit:gov-code-search (orchestrator). Searches govreposcrape for one query variation and fetches GitHub repository pages to extract factual evidence. Returns a JSON payload conforming to arckit-claude/schemas/gov-repo-handoff.schema.json. Not user-invocable — only the or
arckit-gov-code-search-writerWriter subagent invoked by /arckit:gov-code-search (orchestrator). Renders a validated, scored payload into a GCSR artefact under projects/{P}-{NAME}/research/. Has no web/MCP/Agent tools — can only render structured input it is given. Not user-invocable — only the orchestrator dispatches this subag
arckit-gov-code-searchUse this agent when the user wants to search UK government repositories using natural language queries. This agent provides general-purpose semantic search across 24,500+ government repos via govreposcrape. Examples: <example> Context: User wants to find how government teams implemented something us
arckit-gov-landscape-readerReader subagent invoked by /arckit:gov-landscape (orchestrator). Searches govreposcrape and queries vulnerability exposure for one organisation or technology facet, and fetches GitHub repository pages to extract factual evidence. Returns a JSON payload conforming to arckit-claude/schemas/gov-repo-ha
arckit-gov-landscape-writerWriter subagent invoked by /arckit:gov-landscape (orchestrator). Renders a validated, scored payload into a GLND artefact under projects/{P}-{NAME}/research/. Has no web/MCP/Agent tools — can only render structured input it is given. Not user-invocable — only the orchestrator dispatches this subagen
arckit-gov-landscapeUse this agent when the user wants to understand what UK government has built in a domain — mapping organisations, technology patterns, standards, and maturity levels. Examples: <example> Context: User wants to understand the government landscape for a domain user: "/arckit:gov-landscape Map the gov
arckit-gov-reuse-readerReader subagent invoked by /arckit:gov-reuse (orchestrator). Searches govreposcrape and fetches GitHub repository pages to extract reuse- evidence for one capability bucket. Returns a JSON payload conforming to arckit-claude/schemas/gov-reuse-handoff.schema.json. Not user-invocable — only the orches
arckit-gov-reuse-writerWriter subagent invoked by /arckit:gov-reuse (orchestrator). Renders a validated, scored payload into a GOVR artefact under projects/{P}-{NAME}/research/. Spawns one tech-note per Fork/Library candidate. Has no web/MCP/Agent tools — can only render structured input it is given. Not user-invocable —
arckit-gov-reuseUse this agent when the user wants to discover reusable UK government open-source code before building from scratch. This agent searches 24,500+ government repositories via govreposcrape and assesses candidates for reusability. Examples: <example> Context: User has a project and wants to check for e
arckit-grants-readerReader subagent invoked by /arckit:grants (orchestrator). Fetches and extracts factual evidence about UK funding programmes for one funder_category bucket. Returns a JSON payload conforming to arckit-claude/schemas/grants-handoff.schema.json. Not user-invocable — only the orchestrator dispatches thi
arckit-grants-writerWriter subagent invoked by /arckit:grants (orchestrator). Renders a validated, scored payload into a GRNT artefact under projects/{P}-{NAME}/research/. Spawns one tech-note per scored programme. Has no web/MCP/Agent tools — can only render structured input it is given. Not user-invocable — only the
arckit-grantsUse this agent when the user needs to research UK funding opportunities for a project, including government grants (UKRI, Innovate UK, NIHR, DSIT), charitable foundations (Wellcome, Nesta), social impact funding, and accelerator programmes. This agent performs extensive web research autonomously. Ex
arckit-independent-scorerIndependent scorer started by /arckit:score and /arckit:evaluate. Scores a supplier's validated claims against the evaluation criteria without seeing the first evaluator's scores, so the two can be moderated. Everything it needs is in its prompt. Its only tool, Glob, lists file names and can't open
arckit-research-readerReader subagent invoked by /arckit:research (orchestrator). Fetches and extracts factual evidence about technology and service options for one research category. Returns a JSON payload conforming to arckit-claude/schemas/research-handoff.schema.json. Not user-invocable — only the orchestrator dispat
arckit-research-writerWriter subagent invoked by /arckit:research (orchestrator). Renders a validated payload of options, facts and costs into an RSCH artefact under projects/{P}-{NAME}/research/. Spawns one vendor profile per vendor and one tech-note per significant technology finding. Has no web/MCP/Agent tools — can o
arckit-researchUse this agent when the user needs technology and service market research for a project: the options available, checked facts about each, published costs, and UK Government Digital Marketplace search. It reports evidence and makes no build-or-buy decision. This agent performs extensive web research
arckit-score-moderatorNeutral moderator started by /arckit:score and /arckit:evaluate when two evaluators score a criterion differently. Sees both scores and their justifications, labelled A and B with no indication of which came first, plus the criterion's claims and the scale, and decides the final score. Everything it
arckit-scorePrompt for /arckit:score on runtimes with no subagents. Score vendor proposals against evaluation criteria, reading supplier documents in one context. Not for dispatch in Claude Code: run the /arckit:score command, which keeps bids out of the scoring context.
arckit-tenders-readerReader subagent invoked by the /arckit:tenders and /arckit:competitors orchestrator commands. Queries the UK Tenders MCP and extracts factual procurement market evidence (suppliers, buyers, award aggregates, time series, source freshness) for one query scope. Returns a JSON payload conforming to arc
arckit-tenders-writerWriter subagent invoked by the /arckit:tenders orchestrator command. Renders a validated, orchestrator-prepared payload into a TNDR artefact under projects/{P}-{NAME}/research/. Has no web/MCP/Agent tools — can only render structured input it is given. Not user-invocable — only the orchestrator comm
arckit-tendersUse this agent when the user needs UK procurement market intelligence — award-value benchmarks, top suppliers, incumbency and concentration — drawn from the UK Tenders MCP over more than 700,000 UK contracting processes (October 2026). Examples: <example> Context: User wants award-value benchmarks b