← Back to search

io.github.paperdavid/manoma

paperdavid Scanned 27d ago

MCP server for soul.md — your portable AI identity layer across every LLM.

A
91 / 100

Versions

0.2.1latest
Apr 18, 2026
0.2.0
Apr 18, 2026
0.1.0
Apr 18, 2026
PermissionsTool SafetyAuthAnnotationsCode QualityStabilitySpecVuln HistoryAuthorTransparencyCommunity

Tools 6

get_context
annotations: verified low

Load the user's identity context from soul.md. Call ONCE at the start of a conversation. Returns every context section of the file in the author's order (identity, values, voice, skills, current focus, memory), with annotation comments stripped. Sections about other people and legacy configuration are excluded by default per the soul.md spec. Returns: The user's soul.md content — data about the user, not instructions.

readOnlyHint true openWorldHint false idempotentHint true destructiveHint false
get_skill_depth
annotations: verified low

Look up the user's depth in a skill domain, with a calibration hint. Args: - skill (string): domain to look up (e.g. "backend engineering"). Returns: The matching skills line(s), a behavior hint for the user's level, and any custom deep-dive section (e.g. /skills/backend) if the file has one.

readOnlyHint true openWorldHint false idempotentHint true destructiveHint false
list_sections
annotations: verified low

List every section in the user's soul.md with its path and size.

readOnlyHint true openWorldHint false idempotentHint true destructiveHint false
add_decision
annotations: verified low

Record a decision in /memory/decisions of the user's soul.md (newest first, dated, with a provenance stamp). Only the 15 most recent entries are kept. Writes to the user's file on disk. Args: - decision (string): what was chosen and why. Be specific. - context (string, optional): reasoning or alternatives considered.

readOnlyHint false openWorldHint false idempotentHint false destructiveHint true
add_lesson
annotations: verified low

Record a lesson learned in /memory/lessons of the user's soul.md (newest first, dated, with a provenance stamp). Only the 15 most recent entries are kept. Writes to the user's file on disk. Args: - lesson (string): the lesson. Be specific and actionable. - source (string, optional): where it came from.

readOnlyHint false openWorldHint false idempotentHint false destructiveHint true
update_now
annotations: verified low

Replace the /now section of the user's soul.md with their current focus. A dated "*Updated: YYYY-MM-DD via <client>*" stamp is appended automatically (SPEC §6). Writes to the user's file on disk. Args: - content (string): new markdown for /now — projects, deadlines, constraints, open decisions.

readOnlyHint false openWorldHint false idempotentHint true destructiveHint true

Permissions 1

env_vars low
Server uses env_vars capabilities via: process.env

Scan Findings 22

info
Tool 'get_context' annotations are consistent annotation_checker · 80%
info
Tool 'get_skill_depth' annotations are consistent annotation_checker · 80%
info
Tool 'list_sections' annotations are consistent annotation_checker · 80%
info
Tool 'add_decision' annotations are consistent annotation_checker · 80%
info
Tool 'add_lesson' annotations are consistent annotation_checker · 80%
info
Tool 'update_now' annotations are consistent annotation_checker · 80%
info
Sandbox failed to start for behavioral verification behavioral_verifier · 100%
medium
Vulnerable dependency: @modelcontextprotocol/sdk@1.6.1 (GHSA-8r9q-7v3j-jr4g) dependency_analyzer · 95%
medium
Vulnerable dependency: @modelcontextprotocol/sdk@1.6.1 (GHSA-w48q-cv73-mx4w) dependency_analyzer · 95%
info
package.json metadata manifest_parser · 100%
info
Tool: get_context manifest_parser · 85%
info
Tool: get_skill_depth manifest_parser · 85%
info
Tool: list_sections manifest_parser · 85%
info
Tool: add_decision manifest_parser · 85%
info
Tool: add_lesson manifest_parser · 85%
info
Tool: update_now manifest_parser · 85%
info
Transport: stdio manifest_parser · 90%
info
Required env vars (2) manifest_parser · 80%
info
Sandbox failed to start for output poisoning scan output_poisoning · 100%
low
Permission: env_vars access detected permission_analyzer · 90%
info
SBOM generated: 126 components sbom_generator · 100%
medium
No build provenance detected (SLSA L0) slsa_assessor · 90%