Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
JAMBAVAN_ROOTNoProject root to index and serveauto-detect
JAMBAVAN_DEV_MODENoDefault Vibhishana Niti level (lite / full / ultra)full
JAMBAVAN_ALLOW_BASHNo1 registers bashoff
JAMBAVAN_ALLOW_WRITENo1 registers write_file + patch_fileoff
JAMBAVAN_TOKEN_BUDGETNoMax tokens in jambavan_context output8000
JAMBAVAN_ALLOW_SECRETSNo1 lets file tools touch secret-looking filesoff
JAMBAVAN_MAX_READ_BYTESNoMax file size read_file will load5242880
JAMBAVAN_BASH_INHERIT_ENVNo1 passes the full host env to bash (default: minimal env)off
JAMBAVAN_MAX_OUTPUT_CHARSNoGlobal cap on any tool's returned output100000
JAMBAVAN_ALLOW_OUTSIDE_ROOTNo1 lets tools escape the project root (trusted local use only)off

Capabilities

Features and capabilities supported by this server

CapabilityDetails
tools
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
jambavan_awakenA

Return the Jambavan operating protocol: recall memory, index/watch code, retrieve context before edits, patch surgically, run checks, and store durable decisions. Call once at the start of every host session. Includes recent project memories by default.

jambavan_indexA

Build or refresh the Jambavan codebase index. Parses source files with a tree-sitter AST extractor and stores symbols in a local SQLite database. Run once per project, then incrementally on file changes. Returns indexing statistics (files processed, symbols extracted, duration).

jambavan_contextA

Search the Jambavan index for code symbols and snippets most relevant to a query. Returns a ranked, token-budgeted context block of matching functions, classes, and types. Inject this block into your prompt to give the model precise, token-efficient codebase knowledge. Much cheaper than reading whole files — only the relevant symbol bodies are returned. Automatically includes top project-memory matches and a bounded set of extracted callers/callees under the same token budget. Options: compress_prose shrinks comments for extra budget; include_diff adds recent git changes; include_tests shows test coverage.

jambavan_watchA

Control the live file watcher that keeps the index in sync as you edit code. Actions: "start" — begin watching (index must exist first); "stop" — stop watching; "status" — show watcher state. While running, supported non-ignored source-file changes trigger incremental re-indexing (no full rescan). Use "start" after jambavan_index, then forget about it — the index stays fresh automatically.

jambavan_vibhishana_nitiA

Activate Vibhishana Niti in the current session: efficient, truthful senior-dev rules. Returns a concise ruleset: YAGNI-first, stdlib before dependencies, shortest working diff, root-cause fixes. Levels: "lite" — build what's asked, name the leaner option; "full" — the default ladder; "ultra" — deletion extremist. Inject the returned text into your system prompt or next message.

jambavan_rin_mochanA

Harvest every rin marker (lines marked with a ceiling and upgrade path) from the project into a debt ledger. Groups findings by file and flags any marker with no upgrade trigger — those are the ones that silently rot. Read-only. Call before a release or refactor sprint.

jambavan_graph_reportA

Build a lightweight knowledge graph from the current code index and return hub nodes plus edge confidence notes. Call jambavan_index first. Inferred ambiguous-name edges are excluded unless include_inferred=true. Defaults to the first 5000 indexed symbols; higher symbol_limit costs more.

jambavan_graph_queryA

Query the current knowledge graph: find matching nodes and BFS neighbors within a token budget. Call jambavan_index first. Uses extracted edges by default; set include_inferred=true to include ambiguous same-name candidates.

jambavan_graph_pathA

Find the shortest path between two graph nodes/symbols using BFS over extracted edges. Call jambavan_index first. Set include_inferred=true to opt into ambiguous same-name edges.

jambavan_diagnosticsA

Show which languages are backed by tree-sitter (full AST) vs regex fallback, plus current index statistics. Useful for verifying the parser setup after a fresh install.

jambavan_doctorA

One-shot environment health check: root detection source, parser backends, write/bash tool gates, token budget, memory dir writability, .gitignore/CI presence, and index/watcher status. Call this first when something feels off (e.g. context results look like the wrong project) — it catches the most common cause: the MCP host resolving the wrong project root.

jambavan_memory_storeA

Persist a memory as an Open Knowledge Format (OKF) concept document. Memories are markdown files with YAML frontmatter — human-readable and portable. Each memory has a title, body (markdown), optional tags, and a scope (e.g. project name). Memories with the same title in the same scope are overwritten (idempotent). Returns the OKF concept ID (scope/slug) of the stored document.

jambavan_memory_getA

Fetch one complete local memory or MemPalace drawer by ID. MemPalace access is explicit and read-only.

jambavan_memory_search

Search Jambavan memories using BM25 by default. Set provider=mempalace for explicit read-only semantic search, or provider=all for separate provider sections. Jambavan scope/collection filters map to MemPalace wing/room.

jambavan_memory_recallA

Recall up to limit active Jambavan memories, newest first (default: 20). Set provider=mempalace for an explicit read-only drawer listing, or provider=all for separate sections. Ordinary recall never starts MemPalace.

jambavan_memory_mine_sessionA

Mine durable facts, decisions, TODOs, and constraints from a pasted session transcript or log. Stores each extracted item as an OKF memory in the requested scope. This is deterministic text mining — no summarization or external service.

jambavan_memory_invalidateA

Mark a memory as invalidated/superseded without deleting the OKF document. Use when a durable fact changed and you want temporal history instead of silent overwrite.

jambavan_memory_delete

Delete a memory by its OKF concept ID (scope/slug), or delete all memories in a scope. Provide id to delete a single memory. Provide scope with delete_scope: true to wipe all memories in that scope. Returns confirmation with the count of deleted memories.

jambavan_memory_status

Show active Jambavan counts by scope and collection by default. Set provider=mempalace or provider=all to explicitly include read-only MemPalace status and taxonomy.

jambavan_failure_store

Store a structured failure record in the memory palace. Use after a command/approach fails to prevent repeating the same mistake in future sessions. Stores: command, symptom, attempted fix, root cause, resolution, status, and advice on what NOT to retry.

jambavan_failure_search

Search past failure records for a command, error, or symptom. Call BEFORE retrying a failing command to check if it was already diagnosed. Returns structured failure records with status and resolution info.

jambavan_session_exportA

Export current session context as a portable handoff document. Contains: recent memories, rin debt markers, and git status. Use to transfer context to a new session, different host, or a colleague. The output is self-contained markdown — paste it into a new session to resume.

jambavan_session_importA

Import a session handoff document into the memory store. Parses the handoff markdown and stores key items as memories if they are not already present. Idempotent — re-importing the same document does not create duplicates.

jambavan_review_packA

Assemble a review pack for the current branch vs a base branch: touched files, the symbols changed in them, their callers (via the knowledge graph), related tests, past failure records mentioning the same files, and risk flags (open rin debt, no matching test). Call before opening/updating a PR, or whenever asked "what changed" / "review this branch". Requires jambavan_index to have been run at least once for symbol/caller/test/risk analysis.

jambavan_impactA

Analyze changed symbols against inbound extracted graph edges and associated tests. Reports bounded transitive callers, test coverage, and explicit graph incompleteness. Use before a risky refactor or PR review; call jambavan_index first.

jambavan_mool_kaaran

Return a structured root-cause investigation protocol. Call BEFORE attempting to fix any bug, test failure, or unexpected behavior. Prevents guess-and-check thrashing by enforcing observe → compare → hypothesize → fix phases. If attempts_so_far >= 3, returns an escalation protocol (architecture problem likely).

jambavan_praman

Return a verification gate protocol demanding fresh evidence before claiming completion. Call BEFORE asserting that tests pass, a build succeeds, a bug is fixed, or requirements are met. Forces: identify proof command → run it → paste output → only then make the claim. Types: tests, build, fix, requirements, general.

jambavan_yukti

Return a strategic approach protocol for planning multi-step work. Call BEFORE starting implementation of any non-trivial task. Returns phased instructions scaled to task size: small (just do it), medium (2-3 approaches + sequence), large (decompose + checkpoint). Auto-infers scale from task description if not provided.

jambavan_vibhaajan

Return a parallel work decomposition protocol for splitting a task into independent units. Call when a task has sub-units that can proceed independently (different files, no shared state). Returns: boundary identification, independence verification, contract definition, merge sequencing. Works for both multi-agent parallelism and solo context-switching with clean commits.

root_cause

Functional alias for jambavan_mool_kaaran. Return a structured root-cause investigation protocol. Call BEFORE attempting to fix any bug, test failure, or unexpected behavior. Prevents guess-and-check thrashing by enforcing observe → compare → hypothesize → fix phases. If attempts_so_far >= 3, returns an escalation protocol (architecture problem likely).

verify_gate

Functional alias for jambavan_praman. Return a verification gate protocol demanding fresh evidence before claiming completion. Call BEFORE asserting that tests pass, a build succeeds, a bug is fixed, or requirements are met. Forces: identify proof command → run it → paste output → only then make the claim. Types: tests, build, fix, requirements, general.

strategy_planA

Functional alias for jambavan_yukti. Return a strategic approach protocol for planning multi-step work. Call BEFORE starting implementation of any non-trivial task. Returns phased instructions scaled to task size: small (just do it), medium (2-3 approaches + sequence), large (decompose + checkpoint). Auto-infers scale from task description if not provided.

decompose_taskA

Functional alias for jambavan_vibhaajan. Return a parallel work decomposition protocol for splitting a task into independent units. Call when a task has sub-units that can proceed independently (different files, no shared state). Returns: boundary identification, independence verification, contract definition, merge sequencing. Works for both multi-agent parallelism and solo context-switching with clean commits.

dev_rulesA

Functional alias for jambavan_vibhishana_niti. Activate Vibhishana Niti in the current session: efficient, truthful senior-dev rules. Returns a concise ruleset: YAGNI-first, stdlib before dependencies, shortest working diff, root-cause fixes. Levels: "lite" — build what's asked, name the leaner option; "full" — the default ladder; "ultra" — deletion extremist. Inject the returned text into your system prompt or next message.

debt_ledgerA

Functional alias for jambavan_rin_mochan. Harvest every rin marker (lines marked with a ceiling and upgrade path) from the project into a debt ledger. Groups findings by file and flags any marker with no upgrade trigger — those are the ones that silently rot. Read-only. Call before a release or refactor sprint.

read_file

Read the contents of a project file. Paths are confined to the project root unless JAMBAVAN_ALLOW_OUTSIDE_ROOT=1.

searchA

Search project files using ripgrep, with grep fallback. Fast regex search. Paths are confined to the project root unless JAMBAVAN_ALLOW_OUTSIDE_ROOT=1.

list_filesA

List project files in a directory. Paths are confined to the project root unless JAMBAVAN_ALLOW_OUTSIDE_ROOT=1.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/beingmartinbmc/jambavan'

If you have feedback or need assistance with the MCP directory API, please join our Discord server