Recallforge MCP Server
recallforge
PyPI
v0.3.1
Published by brianmeyer — no publish provenance, so origin is unverified, but the source is public: the repository link below is self-declared yet readable, so you can inspect the code before adopting it.
The grade answers one question — how safe is this server for you to adopt — so it is computed in two auditable stages. Nothing below is an opinion or an LLM's guess; every line is a real term the deterministic engine applied, and the same input always yields the same number.
1. Threat score — 100 − 23.1 = 77. What the published surface and source actually contain:
| Points | What was found | Category |
|---|---|---|
| −23.1 | Dynamic evaluation of a non-literal value ×2 MTC-SRC-010 | injection |
2. Client adoption risk — 77 − 11 = 66. Three small, subtract-only factors that reflect your risk in adopting it — a clean scan proves less on a powerful, unverified or barely-inspectable package, so the grade says so plainly:
| Points | Adoption-risk factor |
|---|---|
| −10 | capability blast radius (critical) — client exposure if the model is manipulated |
| −1 | publisher verification (public source) — no provenance, but the source is public and inspectable |
| ✓ | inspection depth (source) — how much of the target the scan could see |
Capability observations and info notes are shown under Findings but never scored.
Open any row's finding below for the file, line and evidence behind a deduction.
Findings 31
high Dynamic evaluation of a non-literal value (src/recallforge/backends/__init__.py)MTC-SRC-010
In the server's implementation (`src/recallforge/backends/__init__.py:53`): Evaluating a runtime value as code (rather than a fixed literal) executes whatever reaches it — a direct RCE primitive, and almost never necessary in legitimate code. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: eval(
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/recallforge/backends/__init__.py
high Dynamic evaluation of a non-literal value (src/recallforge/backends/mlx_backend.py)MTC-SRC-010
In the server's implementation (`src/recallforge/backends/mlx_backend.py:530`): Evaluating a runtime value as code (rather than a fixed literal) executes whatever reaches it — a direct RCE primitive, and almost never necessary in legitimate code. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: eval(
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/recallforge/backends/mlx_backend.py
high Dynamic code execution in server code (src/recallforge/backends/__init__.py)MTC-SRC-001
In the server's implementation (`src/recallforge/backends/__init__.py:53`): Evaluating strings as code is the most direct RCE primitive; if any tool input reaches it, the server executes attacker-chosen code. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: eval(
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/recallforge/backends/__init__.py
high Dynamic code execution in server code (src/recallforge/backends/mlx_backend.py)MTC-SRC-001
In the server's implementation (`src/recallforge/backends/mlx_backend.py:530`): Evaluating strings as code is the most direct RCE primitive; if any tool input reaches it, the server executes attacker-chosen code. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: eval(
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/recallforge/backends/mlx_backend.py
high Shell/command execution in server code (src/recallforge/backends/__init__.py)MTC-SRC-002
In the server's implementation (`src/recallforge/backends/__init__.py:57`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: subprocess.run(
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/recallforge/backends/__init__.py
high Shell/command execution in server code (src/recallforge/search.py)MTC-SRC-002
In the server's implementation (`src/recallforge/search.py:700`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: subprocess.run(
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/recallforge/search.py
high Shell/command execution in server code (src/recallforge/storage/indexing_ops.py)MTC-SRC-002
In the server's implementation (`src/recallforge/storage/indexing_ops.py:672`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: subprocess.run(
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/recallforge/storage/indexing_ops.py
high Shell/command execution in server code (src/recallforge/video.py)MTC-SRC-002
In the server's implementation (`src/recallforge/video.py:229`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: subprocess.run(
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/recallforge/video.py
low Unconstrained path parameter "image_path" on "search"MTC-CAP-008
Tool "search" takes a path parameter "image_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool search · inputSchema.properties.image_path
low Unconstrained path parameter "video_path" on "search"MTC-CAP-008
Tool "search" takes a path parameter "video_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool search · inputSchema.properties.video_path
low Unconstrained path parameter "file_path" on "search"MTC-CAP-008
Tool "search" takes a path parameter "file_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool search · inputSchema.properties.file_path
low Unconstrained path parameter "image_path" on "search_vec"MTC-CAP-008
Tool "search_vec" takes a path parameter "image_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool search_vec · inputSchema.properties.image_path
low Unconstrained path parameter "video_path" on "search_vec"MTC-CAP-008
Tool "search_vec" takes a path parameter "video_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool search_vec · inputSchema.properties.video_path
low Unconstrained path parameter "file_path" on "search_vec"MTC-CAP-008
Tool "search_vec" takes a path parameter "file_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool search_vec · inputSchema.properties.file_path
low Unconstrained path parameter "path" on "ingest"MTC-CAP-008
Tool "ingest" takes a path parameter "path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool ingest · inputSchema.properties.path
low Unconstrained path parameter "file_path" on "ingest"MTC-CAP-008
Tool "ingest" takes a path parameter "file_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool ingest · inputSchema.properties.file_path
low Unconstrained path parameter "folder_path" on "ingest"MTC-CAP-008
Tool "ingest" takes a path parameter "folder_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool ingest · inputSchema.properties.folder_path
low Unconstrained path parameter "max_file_size_mb" on "ingest"MTC-CAP-008
Tool "ingest" takes a path parameter "max_file_size_mb" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool ingest · inputSchema.properties.max_file_size_mb
low Unconstrained path parameter "path" on "index_document"MTC-CAP-008
Tool "index_document" takes a path parameter "path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool index_document · inputSchema.properties.path
low Unconstrained path parameter "path" on "index_image"MTC-CAP-008
Tool "index_image" takes a path parameter "path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool index_image · inputSchema.properties.path
low Unconstrained path parameter "path" on "index_audio"MTC-CAP-008
Tool "index_audio" takes a path parameter "path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool index_audio · inputSchema.properties.path
low Unconstrained path parameter "path" on "memory_add"MTC-CAP-008
Tool "memory_add" takes a path parameter "path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool memory_add · inputSchema.properties.path
low Unconstrained path parameter "path" on "memory_add_conversation"MTC-CAP-008
Tool "memory_add_conversation" takes a path parameter "path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool memory_add_conversation · inputSchema.properties.path
low Unconstrained path parameter "path" on "memory_update"MTC-CAP-008
Tool "memory_update" takes a path parameter "path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool memory_update · inputSchema.properties.path
low Unconstrained path parameter "path" on "memory_delete"MTC-CAP-008
Tool "memory_delete" takes a path parameter "path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool memory_delete · inputSchema.properties.path
low Unconstrained path parameter "path" on "memory_get"MTC-CAP-008
Tool "memory_get" takes a path parameter "path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool memory_get · inputSchema.properties.path
low Unconstrained path parameter "path" on "memory_graph_entities"MTC-CAP-008
Tool "memory_graph_entities" takes a path parameter "path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool memory_graph_entities · inputSchema.properties.path
low Unconstrained path parameter "path" on "memory_graph_related"MTC-CAP-008
Tool "memory_graph_related" takes a path parameter "path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool memory_graph_related · inputSchema.properties.path
low Unconstrained path parameter "image_path" on "explain_results"MTC-CAP-008
Tool "explain_results" takes a path parameter "image_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool explain_results · inputSchema.properties.image_path
low Unconstrained path parameter "video_path" on "explain_results"MTC-CAP-008
Tool "explain_results" takes a path parameter "video_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool explain_results · inputSchema.properties.video_path
low Unconstrained path parameter "file_path" on "explain_results"MTC-CAP-008
Tool "explain_results" takes a path parameter "file_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool explain_results · inputSchema.properties.file_path
What this scan could not see
- Tools were statically extracted from the published source (26 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>".
Versions
Scan history per published version. The engine is deterministic — the same version always yields the same score, so a changed score means the package itself changed.
| Version | Score | Findings | Engine | Scanned |
|---|---|---|---|---|
v0.3.1 latest |
D 66/100 | 31 | 1.7.0 | 2026-07-23 |
Embed this score
Show this server's live Trust Score in your README, docs or website. The badge is served straight from the registry and updates automatically after every rescan — no API key needed. It links back to this page, so anyone who sees the grade can also read the findings behind it instead of taking a number on faith.
Verify this score yourself
The score above is reproducible: the same package version always yields the same result. Run it locally or over the free API — no account, no LLM, fully deterministic.
npx mcptrustchecker scan recallforge --online --registry pypiMore in Developer Tools
Security scan results for the 1stay MCP server.
Security scan results for the Adguard Home MCP server.
Security scan results for the Ado Browser MCP server.
Security scan results for the Aemet MCP server.
Security scan results for the Agent MCP server.
Security scan results for the Agent Recorder MCP server.