Name: Axiom Meridian
Author: abeljumanuel

Axiom Meridian MCP

Serves only the team rules and lessons learned relevant to a project. Code drifts. Meridian doesn't.

Grade A Local · stdio No auth needed 28 tools v0.1.0

No reviews yet by abeljumanuelPyPI 343 Runs on your machine, nothing hosted

97/100

trust score

downloads a day

local

runs on your machine

downloads in 30d

What it does

Serves only the team rules and lessons learned relevant to a project. Code drifts. Meridian doesn't.

Quickstart

# 1 — add it to Claude Code
claude mcp add --scope user axiom-meridian-mcp '--' uvx --from axiom-meridian meridian-mcp

# 2 — ask your agent for something these tools do
#   index_rules_from_markdown, index_lessons_from_markdown, convert_to_atomic_format and 25 more

Axiom Meridian costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.

Publishernot claimed

abeljumanuel1 server

1 category

All 1 server

Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.

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 Axiom Meridian 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. Its scan result is on the Security tab — read the surface 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 uv (docs.astral.sh/uv): uvx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. The '--' is in quotes so that PowerShell passes it on.

claude mcp add --scope user axiom-meridian-mcp '--' uvx --from axiom-meridian meridian-mcp

Reconnect, 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 package: 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

Nothing is required, but it can take KNOWLEDGE_BASE_PATH (Knowledge base root (Markdown files + SQLite index). Defaults to an OS-specific per-user data directory.), MERIDIAN_ACCESS_LEVEL (Which tools may run: read, analyze, or write (write is needed to approve proposals).) if you want to set them.

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

28 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.

Only tools that write or reach the network

Index rules from a markdown file directly into SQLite.

Parameter	Type	Required	Notes
filepath	string	yes	Path to the source .md file containing rule blocks.
default_scope_id	string	yes	Scope assigned to blocks that don't declare their own Scope: field (e.g. 'global-java', 'project-<id>').
mode	string	no	'atomic': parse canonical ## RN-XXX-NNN blocks (the {TECH} segment is optional — plain ## RN-NNN is also accepted) and upsert them directly into rules/rule_history. 'legacy': parse old ### blocks into

Index lessons from a markdown file directly into SQLite.

Parameter	Type	Required	Notes
filepath	string	yes	Path to the source .md file containing lesson blocks.
default_scope_id	string	yes	Scope assigned to blocks that don't declare their own Scope: field (e.g. 'global-java', 'project-<id>').
mode	string	no	'atomic': parse canonical ## LL-XXX-NNN blocks (the {TECH} segment is optional — plain ## LL-NNN is also accepted) and upsert them directly into lessons/lesson_history. 'legacy': parse old ### blocks

Convert a legacy markdown file into atomic-format pending_proposals, one per legacy block, with best-effort scope inference from each block's text. Never writes .md files directly — always stages for human review via approve_proposal.

Parameter	Type	Required	Notes
filepath	string	yes	Path to the legacy (### -style) markdown file to convert.
default_scope_id	string	yes	Fallback scope used when scope inference from the block's text doesn't match a known technology keyword.
doc_type	string	yes	'rules' or 'lessons' — which legacy parser and destination to use.

List pending proposals for human review, optionally filtered by project, type, or status.

Parameter	Type	Required	Notes
project_id	string	no	Filter by project/scope id (with or without the 'project-' prefix).
type	string	no	Filter by proposal type: 'rule', 'lesson', or 'update'.
status	string	no	Filter by status: 'pending', 'approved', or 'rejected'.

Approve a pending proposal. For type='rule'/'lesson' this assigns a new RN-/LL- code, appends the atomic block to the scope's .md file, and inserts the corresponding row; for type='update' it splices the target's existing block in place instead. This is the single choke point that writes proposal…

Parameter	Type	Required	Notes
proposal_id	string	yes	The pending_proposals id to approve.

Update the proposed_text and/or metadata of a still-pending proposal;

Parameter	Type	Required	Notes
proposal_id	string	yes	The pending_proposals id to edit.
new_text	string	yes	Replacement proposed_text (private tags are stripped before storage).
metadata	string	no	JSON object merged into the proposal's existing metadata.

Reject a pending proposal. No side effects besides updating the record.

Parameter	Type	Required	Notes
proposal_id	string	yes	The pending_proposals id to reject.
reason	string	no	Optional human-readable rejection reason.

Promote a rule to a more general scope: marks the original as deprecated (in both SQLite and its source .md file) and records a PROMOTED rule_history entry. Does not create the rule in the new scope — that requires a separate proposal.

Parameter	Type	Required	Notes
rule_id	string	yes	The RN-* id of the rule to promote.
new_scope_id	string	yes	Destination scope to promote the rule into.

Create a proposal to deprecate a rule or lesson on its own, without changing its scope (unlike promote_rule). Accepts both RN-* and LL-* ids. Only creates a pending_proposals row (type='deprecate') — no immediate effect; approve_proposal performs the actual deprecation after human review.

Parameter	Type	Required	Notes
rule_id	string	yes	The RN-* or LL-* id of the rule/lesson to deprecate.
reason	string	yes	Why this rule/lesson is being deprecated.
superseded_by	string	no	Optional RN-/LL- id of the active rule/lesson that replaces this one.

Create a new project scope and its knowledge base directories: a 'project-<id>' scope under parent_scope, plus knowledge-base/projects/ and lessons/projects/ directories with a placeholder rules .md file.

Parameter	Type	Required	Notes
project_id	string	yes	New project identifier (the tool creates scope 'project-<id>').
name	string	no	Human-readable project name; defaults to project_id.
parent_scope	string	no	Existing scope this project inherits from in the scope hierarchy.

Generate or update ChromaDB embeddings for active rules/lessons whose embedding_id is still NULL — an incremental backfill, not a full re-embed.

Parameter	Type	Required	Notes
scope_id	string	no	Restrict to one scope; omit to process every scope.

Generate a Claude Code skill for a project from its resolved rules:

Parameter	Type	Required	Notes
project_id	string	yes	Project scope to generate skill files for.
project_path	string	no	Filesystem path to the target project; falls back to MERIDIAN_PROJECT_PATH if omitted.

Query active rules visible to a project across its full scope hierarchy (most-specific scope first), via plain SQL filtering or — when query_text is set and embeddings exist — semantic search.

Parameter	Type	Required	Notes
project_id	string	yes	Project id (with or without the 'project-' prefix).
category	string	no	Filter by rule category, e.g. 'quality', 'security', 'architecture'.
severity	string	no	Filter by severity: 'low' \| 'medium' \| 'high' \| 'critical'.
tags	string	no	Comma-separated tags; matched by exact tag equality, not substring.
query_text	string	no	Free-text query. When set and ChromaDB has embeddings, enables semantic (RAG) search instead of plain SQL filtering.
format	string	no	Output format: 'json' or 'toon' (compact tabular).
detail	string	no	'summary': metadata only. 'full': also includes rule text (may return STALE_INDEX per row if its source file changed since indexing).
include_deprecated	boolean	no	Include deprecated rules. Only affects plain SQL filtering — deprecated rules are removed from the semantic (query_text) index on deprecation, so this does not revive them there.

Query active lessons visible to a project across its full scope hierarchy (most-specific scope first), via plain SQL filtering or — when query_text is set and embeddings exist — semantic search.

Parameter	Type	Required	Notes
project_id	string	yes	Project id (with or without the 'project-' prefix).
area	string	no	Filter by area_affected, e.g. 'deployment', 'performance', 'security'.
tags	string	no	Comma-separated tags; matched by exact tag equality, not substring.
query_text	string	no	Free-text query. When set and ChromaDB has embeddings, enables semantic (RAG) search instead of plain SQL filtering.
format	string	no	Output format: 'json' or 'toon' (compact tabular).
detail	string	no	'summary': metadata only. 'full': also includes what_happened text (may return STALE_INDEX per row if its source file changed since indexing).
include_deprecated	boolean	no	Include deprecated lessons. Only affects plain SQL filtering — deprecated lessons are removed from the semantic (query_text) index on deprecation, so this does not revive them there.

Return a rule's full text, its complete rule_history, and any lessons linked to it.

Parameter	Type	Required	Notes
rule_id	string	yes	The RN-* id of the rule.

Return a rule's chronological history and linked-lesson summary without loading full text — lighter than get_rule_context.

Parameter	Type	Required	Notes
rule_id	string	yes	The RN-* id of the rule.

Resolve and cache a project's effective scope chain (most-specific first) with each scope's dynamic attributes (ADR-002).

Parameter	Type	Required	Notes
project_id	string	yes	Project id (with or without the 'project-' prefix).

Report this running server process's version/commit, and whether the code on disk has changed since this process started — Python doesn't hot-reload, so an editable install that was git pulled without restarting the server keeps serving the old code with no other visible sign of it (Hallazgo 7).

No parameter schema on file.

Return the chronology of rule_history changes for a project or scope, plus any rules currently deprecated in that scope.

Parameter	Type	Required	Notes
project_id	string	no	Project id to scope the log to (checked before scope_id).
scope_id	string	no	Raw scope id to scope the log to, if project_id is omitted.
since	string	no	ISO-8601 timestamp; only include events at or after this time.

Audit a PR diff against a project's active rules and lessons (detail="full"): persists an audit record (SHA-256 diff hash + a snapshot of the rule versions evaluated) and returns the diff plus rules/lessons for the calling LLM to classify.

Parameter	Type	Required	Notes
pr_diff	string	yes	The PR's unified diff text.
project_id	string	yes	Project id whose active rules/lessons to audit against.

Link PR review feedback to the most recent audit_pr record for the same pr_ref/project_id, returning the rules that applied at that time for the calling LLM to classify each feedback point.

Parameter	Type	Required	Notes
feedback_text	string	yes	Free-text PR review feedback to classify.
pr_ref	string	yes	Reference (e.g. PR number/URL) used to find the matching audit_pr call.
project_id	string	yes	Project id the original audit_pr call used.

Check a planned feature description against a project's active rules and lessons (detail="full"); persists a planning_checks record and returns the context for the calling LLM to flag conflicts.

Parameter	Type	Required	Notes
feature_description	string	yes	Free-text description of the feature being planned.
project_id	string	yes	Project id whose active rules/lessons to check against.

Return everything the calling LLM needs to extract candidate rules from a transcript: the clean text, the project's scope chain with attributes, existing rules for duplicate detection, and the canonical rule template. Does not itself create any proposal — call

Parameter	Type	Required	Notes
text	string	yes	Raw transcript text; private tags are stripped automatically.
project_id	string	yes	Project id used to resolve scopes and existing rules.

Return everything the calling LLM needs to extract candidate lessons from a transcript: the clean text, the project's scope chain with attributes, existing lessons for duplicate detection, and the canonical lesson template. Does not itself create any proposal — call

Parameter	Type	Required	Notes
text	string	yes	Raw transcript text; private tags are stripped automatically.
project_id	string	yes	Project id used to resolve scopes and existing lessons.

Validate and persist a pending proposal of type 'rule', 'lesson', or 'update' (private tags stripped from proposed_text and metadata). The body goes in proposed_text and the other block fields in metadata; a full rendered atomic block in proposed_text is split into the two. For type='update',…

Parameter	Type	Required	Notes
type	string	yes	Proposal type: 'rule', 'lesson', or 'update'.
proposed_text	string	yes	The body only: the rule text (goes under Regla:) or what happened (goes under Qué pasó:). Put the other block fields in metadata. A full rendered atomic block is also accepted and split into
suggested_scope_id	string	yes	Scope to attach the proposal to if approved. Ignored for type='update', which inherits the target's own scope.
target_id	string	no	For type='update': the existing RN-/LL- id this proposal will replace.
target_type	string	no	For type='update': 'rule' or 'lesson'; inferred from target_id's prefix if omitted.
suggested_attributes	string	no	JSON array of {"key": ..., "value": ...} dynamic attributes (ADR-002) to attach if approved.
source_type	string	no	Provenance type, e.g. 'transcript' or 'manual'.
source_ref	string	no	Provenance reference, e.g. a meeting id or PR link.
metadata	string	no	JSON object with the block fields other than the body. Rule keys: category, severity (low\|medium\|high\|critical), applies_to, tags (list or comma-separated), source. Lesson keys: project, date_occurred

Return the canonical ## RN-XXX-NNN atomic block template, its field descriptions, and usage instructions.

Parameter	Type	Required	Notes
project_id	string	no	If given, also returns recommended scope_suggestions for this project.

Return the canonical ## LL-XXX-NNN atomic block template, its field descriptions, and usage instructions.

Parameter	Type	Required	Notes
project_id	string	no	If given, also returns recommended scope_suggestions for this project.

Return the combined rule+lesson template guide and workflow

Parameter	Type	Required	Notes
project_id	string	no	If given, also returns recommended scope_suggestions for this project.

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

Nothing here is billable: mcprush charges nothing to install Axiom Meridian or to call it, at any volume.

ATrust score 97 / 100

Release 0.1.0 · scanned 16 Oct 2026

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.

Threat score100

Client score97

Findings7 threat · 0 capability

Capabilitymoderate

Coveragesource

Verificationsource

Surface digestde2eb2396e660749621e948d790bedeaa2d9145989468b80d7b67553df40d267

Score vectorthreat 100 → client 97

Term	Level	What it prices	Points
capability-exposure	moderate	capability blast radius (moderate) — client exposure if the model is manipulated	−3
verification-discount	source	publisher verification (provenance) — cryptographic build provenance ties the artifact to its source	−0
coverage-honesty	source	inspection depth (source) — how much of the target the scan could see	−0

score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.

What the scan could actually read

A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.

Tool surfaceread

Implementation sourceread

Package metadataread

Live transportnot available

Tools were statically extracted from the published source (28 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".

Capability — what it could do if the model were manipulated

Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. moderate is the level these add up to.

Tool	Capability tags	Why the tag was assigned
index_rules_from_markdown	no tags
index_lessons_from_markdown	no tags
convert_to_atomic_format	no tags
list_pending_proposals	no tags
approve_proposal	no tags
edit_proposal	no tags
reject_proposal	no tags
promote_rule	no tags
deprecate_rule	no tags
create_project	no tags
generate_embeddings	no tags
generate_project_skills	no tags
query_rules	no tags
query_lessons	no tags
get_rule_context	no tags
get_rule_timeline	no tags
get_project_scope_resolution	no tags
get_server_info	no tags
get_rule_audit_log	no tags
audit_pr	no tags
analyze_pr_feedback	no tags
check_feature_against_rules	no tags
extract_rules_from_transcript	no tags
extract_lessons_from_transcript	no tags
create_pending_proposal	no tags
get_rule_template	no tags
get_lesson_template	no tags
get_transcription_template	no tags

Toxic-flow graph

The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.

The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.

Supply chain and provenance

Integrityfirst-seen

Surface digestde2eb2396e660749

Scanned2026-10-16

This is the first scan of this surface here, so there is nothing yet to compare it against.

Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.