long-novel-agent-kit
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_agent_templateC | Return host desktop agent workflow instructions for a local platform and write mode. |
| explain_tool_visibilityC | Explain whether a local MCP tool is visible under read-only/writer mode and core/writing/full profile. |
| build_desktop_next_stepC | Build a read-only next-step decision report for a local desktop agent from project state, MCP config, desktop evidence, and author review queue. |
| build_desktop_diagnosticsC | Build a read-only local desktop-agent diagnostics report from project state, MCP config, desktop evidence, and optional pack health. |
| build_agent_startup_promptB | Build a copy-ready read-only startup prompt for a local desktop agent starting or taking over a target chapter. |
| build_author_review_queueB | Build a read-only queue of proposals, conflicts, writer-mode authorization, and chapter-range acceptance items that need author review before durable state changes. |
| list_agent_activityC | List local desktop agent activity and handoff records. |
| build_agent_activity_reportC | Check whether target-visible desktop agent activity is specific enough for handoff. Read-only. |
| get_project_statusC | Summarize whether a .novel-agent project is ready for desktop-agent writing. |
| list_agent_capabilitiesC | List recorded host agent capability declarations. |
| list_factsB | List active or all structured continuity facts, optionally filtered by chapter and kind. |
| build_chapter_contextC | Build chapter-safe context for the target chapter before writing or revising. |
| build_chapter_context_briefA | Build only the compact read-first chapter context brief for small-context desktop agents. |
| prepare_chapter_sessionB | Build a read-only startup packet for a desktop agent before drafting or revising a target chapter. |
| check_write_sessionB | Read-only preflight before writer tools mutate .novel-agent state. Checks project_id, project-state sha256, and optional target chapter context hash from prepare_chapter_session. |
| check_chapterC | Check a chapter file against visible rules, future boundaries, character state, structured facts, timeline order, location, relationship, life state, and continuity obligations; returns issues plus a revision_plan. |
| diff_chapter_contractB | Compare a chapter file with visible required beats, acceptance checks, forbidden moves, and future markers. |
| build_chapter_readinessC | Build a read-only author-facing chapter acceptance readiness report from continuity checks and contract diff, including a revision_plan ordered for desktop agents. |
| build_chapter_revision_promptC | Build a read-only copyable revision prompt from chapter readiness, revision_plan, and fact_context for a local desktop agent. |
| build_chapter_revision_comparisonB | Compare before and after draft readiness for a read-only revision report with resolved, new, and remaining issues. |
| build_chapter_delivery_packetB | Build a read-only author delivery packet that combines draft metadata, chapter readiness, handoff report, gates, and post-acceptance commands. |
| build_chapter_acceptance_planB | Build a read-only post-acceptance write plan for one draft chapter, including pre-write review, author-confirmation gates, write-session guard guidance, and ordered writer commands. |
| build_proposal_templateC | Build an empty post-write proposal template with chapter context and evidence requirements. |
| get_proposal_schemaA | Return the JSON Schema for post-write proposal updates. |
| validate_proposalB | Validate a post-write update proposal for evidence and conflicts before applying it. |
| diff_proposalC | Preview which facts, characters, debts, and handoff fields a proposal would change. |
| build_proposal_reviewC | Build a read-only author review packet for a saved or inline post-write proposal. |
| build_proposal_readinessC | Build a read-only readiness gate for a saved or inline post-write proposal before durable apply. |
| list_open_threadsC | List visible unresolved plot debts, chapter contracts, conflicts, and the latest handoff. |
| build_handoff_reportB | Build a handoff report and next-agent brief, including required workflow, startup commands, write boundaries, and pending continuity work. |
| doctorC | Check kit installation and optional project .novel-agent state health. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 31 tools
Most tools have distinct purposes, but some overlap exists (e.g., build_agent_activity_report vs list_agent_activity, build_chapter_context vs build_chapter_context_brief). Descriptions help clarify, but agents may still confuse similar-sounding tools.
Tools use a mix of verb prefixes (build_, check_, diff_, list_, get_, etc.) without a single consistent pattern. While each subgroup is internally consistent, the overall naming lacks predictability.
31 tools is on the high side for a single server, but the domain (novel writing agent with chapter management, proposals, handoffs) is complex enough to justify the count. Some tools might be merged.
The tool set covers the core workflow: chapter readiness, proposal management, handoffs, diagnostics, and pre-write sessions. Minor gaps exist (e.g., no direct tool for writing content, but that is delegated to the agent via prompts).