Talamus MCP Server
talamus
PyPI
v1.1.1
Published by ampres-ai — 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 − 24.4 = 76. What the published surface and source actually contain:
| Points | What was found | Category |
|---|---|---|
| −22 | Hardcoded AWS access key id in server code MTC-SRC-008 | exfiltration |
| −1.2 | Authority framing in tool metadata MTC-INJ-AUTH-1 | injection |
| −1.2 | Excessive ALL-CAPS emphasis in metadata MTC-INJ-CAPS | injection |
2. Client adoption risk — 76 − 7 = 69. 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 |
|---|---|
| −6 | capability blast radius (high) — 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 |
Grade capped: 1 confirmed high finding → grade capped at C. A hard gate overrides the point total — no amount of clean surface buys back a confirmed catastrophe.
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 13
high Reads a sensitive credential path or dumps the environment (src/talamus/redact.py)MTC-SRC-006
In the server's implementation (`src/talamus/redact.py:23`): Reading private keys / cloud credentials, or serializing the whole environment, is a sensitive-data source that becomes exfiltration when combined with any egress. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: .netrc
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/talamus/redact.py
high Reads a sensitive credential path or dumps the environment (src/talamus/vault_import.py)MTC-SRC-006
In the server's implementation (`src/talamus/vault_import.py:34`): Reading private keys / cloud credentials, or serializing the whole environment, is a sensitive-data source that becomes exfiltration when combined with any egress. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: ~/.ssh
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/talamus/vault_import.py
high Reads a sensitive credential path or dumps the environment (tests/test_talamus_redact.py)MTC-SRC-006
In the server's implementation (`tests/test_talamus_redact.py:9`): Reading private keys / cloud credentials, or serializing the whole environment, is a sensitive-data source that becomes exfiltration when combined with any egress. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: id_rsa
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 tests/test_talamus_redact.py
high Hardcoded AWS access key id in server code (tests/test_talamus_redact.py)MTC-SRC-008
A live-looking AWS access key id is hardcoded in `tests/test_talamus_redact.py:19`. Secrets in source ship to everyone who installs the package and are a direct credential leak.
Evidence: AWS access key id: AKIA…(redacted)
Fix: Remove the secret, rotate it, and load credentials from the environment or a secret store.
Location: server tests/test_talamus_redact.py
high Shell/command execution in server code (src/talamus/adapters/llm.py)MTC-SRC-002
In the server's implementation (`src/talamus/adapters/llm.py:117`): 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/talamus/adapters/llm.py
high Shell/command execution in server code (src/talamus/bench.py)MTC-SRC-002
In the server's implementation (`src/talamus/bench.py:133`): 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/talamus/bench.py
high Shell/command execution in server code (src/talamus/cli/lifecycle.py)MTC-SRC-002
In the server's implementation (`src/talamus/cli/lifecycle.py:303`): 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/talamus/cli/lifecycle.py
high Shell/command execution in server code (src/talamus/scan.py)MTC-SRC-002
In the server's implementation (`src/talamus/scan.py:110`): 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/talamus/scan.py
high Shell/command execution in server code (src/talamus/services/integrations.py)MTC-SRC-002
In the server's implementation (`src/talamus/services/integrations.py:150`): 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/talamus/services/integrations.py
high Shell/command execution in server code (tests/test_benchmarks_one_screen.py)MTC-SRC-002
In the server's implementation (`tests/test_benchmarks_one_screen.py:19`): 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 tests/test_benchmarks_one_screen.py
high Shell/command execution in server code (tests/test_mcpb_package.py)MTC-SRC-002
In the server's implementation (`tests/test_mcpb_package.py:84`): 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 tests/test_mcpb_package.py
low Authority framing in tool metadataMTC-INJ-AUTH-1
Authority framing in tool metadata detected in the description of tool "remember". Instruction-like content in tool metadata is executed by the model, not the human, and is the primary tool-poisoning vector.
Evidence: important
Fix: Tool descriptions should describe behavior, not instruct the assistant. Treat imperative / secrecy / sequencing language in metadata as hostile.
Location: tool remember · description
low Excessive ALL-CAPS emphasis in metadataMTC-INJ-CAPS
A run of shouted words is a common way to give injected instructions false authority.
Evidence: Read the full Markdown content of a Talamus note given its title.
Optional as_of (a date like "2026-01" or "2026-01
Location: tool read_note · description
What this scan could not see
- Tools were statically extracted from the published source (16 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 |
|---|---|---|---|---|
v1.1.1 latest |
D 69/100 | 13 | 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 talamus --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.