Pack local or remote codebases into AI-friendly files that LLMs and coding agents can read or search
Nothing to run from here
Garuda is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
54
tools declared
—
trust score
local
runs on your machine
—
calls last 7d
What it does
Impact analysis, architecture rules and a tamper-evident ledger for AI coding agents.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user --env 'DATABASE_URL=${DATABASE_URL}' --transport stdio garuda-mcp '--' docker run -i --rm --platform linux/amd64 -e DATABASE_URL ghcr.io/myshra777-ai/garuda-mcp:1.9.11
# 2 — ask your agent for something these tools do
# garuda.briefing, garuda.query, garuda.get_lineage and 51 more
It needs DATABASE_URL: this reads it from your environment, so export it first.
Garuda costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publisher
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts Garuda as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. It has not been scanned here, so read the surface, and its source, before you approve it.
Claude Codedeclared by this listing
This is a public server: you run it yourself and this marketplace is not in the path. You need Docker installed and running. Only the required variables are passed into the container. To set an optional one (GARUDA_WORKSPACE), add -e NAME before the image name and give it a value in the entry’s env: an unset -e would clear the default the image sets. Claude Code registers it in one command; --scope user makes it available in every project. Claude Code reads DATABASE_URL (PostgreSQL connection URL for the Garuda graph and ledger) from your environment each time it starts the server, so the value stays out of its config: export it in your shell profile (macOS, Linux), or run setx DATABASE_URL … and open a new terminal (Windows). The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user --env 'DATABASE_URL=${DATABASE_URL}' --transport stdio garuda-mcp '--' docker run -i --rm --platform linux/amd64 -e DATABASE_URL ghcr.io/myshra777-ai/garuda-mcp:1.9.11Reconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it runs from its publisher’s own image: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it asks of you
It reaches a system of yours, so it needs settings only you can give it: DATABASE_URL — PostgreSQL connection URL for the Garuda graph and ledger. You set it in your own client’s config, on your machine — this marketplace never holds it.
What it writes
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
What you approve
54 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
Session-start briefing: workspace state, policy verdict, epistemic counts, trust anchor, scale, top hubs, active policies, open contradictions, and what changed since this agent's last briefing. Read-only. Call this first.
No parameter schema on file.
Query the Garuda knowledge graph (entities, document claims, and decisions) with natural language keywords. Rejects raw SQL statements.
No parameter schema on file.
Get the full governance and AST lineage of a decision UUID, actor name (e.g. 'cli-operator'), or symbol name.
No parameter schema on file.
Detect unresolved contradictions across decisions and ADR document claims in the tenant knowledge graph.
No parameter schema on file.
Find what breaks if a decision UUID or code symbol (e.g. 'AppendLeafAndSealTx') is changed. Computes transitive blast radius and child decision impact.
No parameter schema on file.
Compute the blast radius of a symbol change: which entities consume the target entity, directly or transitively, up to a depth. Read-only. Returns a critical-first list and per-severity counts.
No parameter schema on file.
Propose a new architectural decision (creates a draft).
No parameter schema on file.
List policies registered for a tenant. Read-only.
No parameter schema on file.
Aggregate governance state: top-level decision, active policies, documentation health, blocked_by list, and unified entity/claim counts. Read-only.
No parameter schema on file.
Bidirectional documentation-to-code (doc_to_code) and code-to-documentation (code_to_doc) drift report. Read-only.
No parameter schema on file.
Query document claims matching a subject and/or epistemic status (SUPPORTED, UNVERIFIED, CONTRADICTED). Read-only.
No parameter schema on file.
Dry-run the policy engine against a workspace and return deterministic policy decisions. Does NOT persist or Merkle-anchor. Use 'garuda policy evaluate' from the CLI to anchor a real governance decision.
No parameter schema on file.
Verify the Merkle inclusion proof for a persisted policy evaluation. Returns valid=true if the proof recomputes to the anchored root. Read-only; no side effects.
No parameter schema on file.
Verify the RFC 6962 Merkle inclusion proof or SHA-256 evidence digest for a policy evaluation, evidence block, or decision UUID. Read-only; no side effects.
No parameter schema on file.
List semantic entities in a workspace. Filterable by package and kind. Read-only.
No parameter schema on file.
Inspect one entity with its signature and incoming and outgoing relationships. Read-only.
No parameter schema on file.
List entities that inherit from or embed the subject. Uses INHERITS and EMBEDS claims. Read-only.
No parameter schema on file.
List entities that implement the subject interface. Uses IMPLEMENTS claims. Read-only.
No parameter schema on file.
List every entity connected to the subject in one hop, both incoming and outgoing directions. External stubs excluded by default. Read-only.
No parameter schema on file.
Create a new task and provision source/target agent UUIDs in the workspace so garuda.handoff and garuda.resume can operate on persisted UUIDs.
No parameter schema on file.
Initiate an atomic task handoff between two agents. Creates a checkpoint of the source agent's state, transfers task ownership, and transitions both agents' statuses. Call garuda.create_task first to provision UUIDs.
No parameter schema on file.
Restore the state from an active checkpoint created by garuda.handoff, and mark the checkpoint as consumed. A second resume of the same checkpoint returns {status: 'not_found'}.
No parameter schema on file.
Find entities by exact name, regex name_pattern, kind, package, or file path. Returns status='not_found' with empty results when a filter matches 0 entities. Read-only.
No parameter schema on file.
MANDATORY pre-edit check: call before modifying any file or symbol. Returns transitive blast radius, callers, applicable ADR MUST/MUST_NOT claims, and active policy guardrails.
No parameter schema on file.
MANDATORY post-edit verification gate: call immediately after modifying code. Runs in dry-run mode (side_effects=[]) by default; set save_to_ledger=true to append a Merkle ledger block (side_effects=['merkle_ledger_append']).
No parameter schema on file.
Compare competing undocumented formulas across teams with 0.0-1.0 team weights summing to 1.00, surface criticality tiers (HIGH_SEVERE, MEDIUM, LOW), and explicit mode ('live_ast' vs 'fixture').
No parameter schema on file.
Audit cross-team business logic and metric number discrepancies in the Data & Logic Audit Ledger by tracing AST and SQL lineage before production deployment.
No parameter schema on file.
Record a reconciliation proposal for a flagged cross-team data discrepancy in the Data & Logic Audit Ledger. Requires GARUDA_MCP_APPROVER or CLI co-signing for human approval.
No parameter schema on file.
Legacy alias for garuda.audit_data_discrepancy: audit cross-team business metric number discrepancies in the Data & Logic Audit Ledger.
No parameter schema on file.
Synthesize unorganized repository architecture, surface criticality tiers, and multi-team formula weights into a structured draft Base Rulebook for human stakeholder review and approval.
No parameter schema on file.
Propose or record a decision on a competing multi-team formula variant in PostgreSQL. Requires GARUDA_MCP_APPROVER or CLI co-signing for APPROVED_BY_HUMAN.
No parameter schema on file.
Legacy alias for garuda.reconcile_data_audit: record reconciliation for a flagged cross-team metric discrepancy in the Data & Logic Audit Ledger.
No parameter schema on file.
Propose or approve the synthesized Base Rulebook with an RFC 6962 SHA-256 digest. Requires GARUDA_MCP_APPROVER or CLI co-signing for human approval.
No parameter schema on file.
List or register 3-Tier Surface Criticality classifications (HIGH_SEVERE freeze, MEDIUM trade-off, LOW UI box alignment).
No parameter schema on file.
Retrieve the 4-Tier Document-to-Code & Guardrails Lineage Matrix (Documents -> Normative Rules -> Active Policies -> AST Symbols).
No parameter schema on file.
Ingest Markdown ADRs or design specifications into document_claims and correlate them against workspace AST symbols.
No parameter schema on file.
Run the Garuda System Doctor diagnostic suite across platform, PostgreSQL, Talon contract, and API gateway health.
No parameter schema on file.
Verify the codebase across the 4-target Cross-Platform Build Matrix (linux/amd64, darwin/arm64, windows/amd64, cgo/purego) and return the deterministic BuildContext receipt.
No parameter schema on file.
Run the static code hygiene analyzer on the workspace graph to report advisory unreferenced candidates, duplicate symbol names, and stdlib alternatives with structured file:line locations.
No parameter schema on file.
Reset smoke-test or agent-proposed audit state in the workspace back to clean baseline.
No parameter schema on file.
Provision an isolated sandbox workspace (e.g. 'audit-sandbox') so mutating governance or audit checks can run safely without affecting the default workspace.
No parameter schema on file.
Recompute receipt_hash, graph_fingerprint, and policy_set_version from the current AST and git commit to cryptographically verify a pre-edit or guard receipt.
No parameter schema on file.
Compare the indexed graph snapshot receipt against current HEAD and list active vs superseded policies, commit sync state, and drift deltas.
No parameter schema on file.
Compute diff-level blast radius in one call: returns per-symbol impact with file, line_start, and source snippets for every symbol touched in the git diff.
No parameter schema on file.
Show which unit tests (*_test.go) and runtime traces exercise a symbol or package, explaining exported handler verification status.
No parameter schema on file.
Return the current workspace snapshot ID, commit SHA, dirty-tree status, policy-set version, and freshness verdict (fresh, stale, dirty) in one lightweight call.
No parameter schema on file.
Return the compiler AST or policy derivation path and source snippet for a claim_id or symbol name.
No parameter schema on file.
Return hash-verified source lines around a symbol or file_path:line_start against the active snapshot.
No parameter schema on file.
Return a compact token-saving workspace summary (counts, active policies, and snapshot ID).
No parameter schema on file.
Return the declared owner and team for an entity from the ownership mapping, or UNASSIGNED if undeclared.
No parameter schema on file.
Verify a structured edge (from_symbol, predicate, to_symbol) or prose claim against the indexed graph and return SUPPORTED, CONTRADICTED, or NOT_IN_GRAPH with a cryptographic receipt.
No parameter schema on file.
Alias for garuda.symbol_test_coverage: inspect which unit tests and runtime traces exercise a symbol.
No parameter schema on file.
Export workspace decisions, policy evaluations, and Merkle block anchors as structured JSON or CSV for compliance filing.
No parameter schema on file.
Single-call token-saving pre-edit bundle combining snapshot freshness, talon_pre_edit briefing, test coverage, and declared ownership for a symbol or file.
No parameter schema on file.
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under Apache-2.0
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install Garuda or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
—Grade not published
No rule-by-rule report on file for this release
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
This listing ships as a container image. The scanner reads packages on npm and PyPI and public source; an image is neither, so there is no grade here — not a withheld one, an unmeasured one.
Release history
It installs from a container image rather than from a versioned package, so the command cannot be pinned to a release. The history below is for reading; the current release on file is 1.9.11.
1.9.11
undated
latestNo release note was published with this version.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish Garuda can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.