Propose a patch

Evidence

The sandbox observed nothing beyond the static findings; the proposal cites those.

diff --git a/src/outlook_mcp/toolsets.py b/src/outlook_mcp/toolsets.py
--- a/src/outlook_mcp/toolsets.py
+++ b/src/outlook_mcp/toolsets.py
@@ -11,1 +11,0 @@
-2. **Config-gated toolsets** — the full tool schemas cost chars/4 proxy tokens
@@ -41,1 +40,0 @@
-# feature here (~1.2k proxy tokens, ~8% of the default surface — mostly
@@ -81,1 +79,0 @@
-# feature here (~1.2k proxy tokens, mostly `create_rule`'s argument list), and
diff --git a/src/outlook_mcp/server.py b/src/outlook_mcp/server.py
--- a/src/outlook_mcp/server.py
+++ b/src/outlook_mcp/server.py
@@ -182,1 +182,0 @@
-# re-fetching ~13k chars/4 proxy tokens of schemas. The set is fixed at import —
diff --git a/tests/test_tool_surface_budget.py b/tests/test_tool_surface_budget.py
--- a/tests/test_tool_surface_budget.py
+++ b/tests/test_tool_surface_budget.py
@@ -268,1 +268,0 @@
-f"The 71-tool default surface now costs ~{cost} proxy tokens per turn, over the "
@@ -319,1 +318,0 @@
-f"The rules-enabled surface costs ~{measured['cost']} proxy tokens, over the "
@@ -332,1 +330,0 @@
-proxy tokens on 1.25.1), and the budget measured a different surface
@@ -353,1 +350,0 @@
-f"`prompts/list` now costs ~{cost} proxy tokens. A prompt's body is supposed "
@@ -413,1 +409,0 @@
-# (+79 proxy tokens). All groups: 77 / 12,175. Measured on Python 3.13 with
diff --git a/tests/test_delta_url_validation.py b/tests/test_delta_url_validation.py
--- a/tests/test_delta_url_validation.py
+++ b/tests/test_delta_url_validation.py
@@ -261,1 +261,0 @@
-pytest.param("https://user:[email protected]/x", id="userinfo"),

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.32.0. If the behavior is intended, declare it (host under permissions.network, a line in the README) instead of removing it; the next crawl scores declarations.

Issue / PR body

Title: outlook-mcp: Credential pass-through configuration (KillerSkills verdict caution 58)

## What the honeypot observed
KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.8.1) and scored it **CAUTION 58/100**. Summary: static analysis only — behavior not yet observed.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/clawhub/outlook-mcp/1.32.0

## Findings with a location
- `src/outlook_mcp/toolsets.py:41` · Credential pass-through configuration — src/outlook_mcp/toolsets.py:41 · Credential pass-through configuration — agent relays sensitive credentials
- `src/outlook_mcp/toolsets.py:81` · Credential pass-through configuration — src/outlook_mcp/toolsets.py:81 · Credential pass-through configuration — agent relays sensitive credentials
- `src/outlook_mcp/toolsets.py:11` · Credential pass-through configuration — src/outlook_mcp/toolsets.py:11 · Credential pass-through configuration — agent relays sensitive credentials
- `src/outlook_mcp/server.py:182` · Credential pass-through configuration — src/outlook_mcp/server.py:182 · Credential pass-through configuration — agent relays sensitive credentials
- `tests/test_tool_surface_budget.py:332` · Credential pass-through configuration — tests/test_tool_surface_budget.py:332 · Credential pass-through configuration — agent relays sensitive credentials
- `tests/test_delta_url_validation.py:261` · URL with embedded credentials — tests/test_delta_url_validation.py:261 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_tool_surface_budget.py:353` · Credential pass-through configuration — tests/test_tool_surface_budget.py:353 · Credential pass-through configuration — agent relays sensitive credentials
- `tests/test_tool_surface_budget.py:413` · Credential pass-through configuration — tests/test_tool_surface_budget.py:413 · Credential pass-through configuration — agent relays sensitive credentials
- `tests/test_tool_surface_budget.py:268` · Credential pass-through configuration — tests/test_tool_surface_budget.py:268 · Credential pass-through configuration — agent relays sensitive credentials
- `tests/test_tool_surface_budget.py:319` · Credential pass-through configuration — tests/test_tool_surface_budget.py:319 · Credential pass-through configuration — agent relays sensitive credentials

## Proposed change
The attached patch removes the 10 lines above. If the behavior is intended, the fix is to declare it instead: list the host under `permissions.network` (or the MCP manifest) and say in the README why the package needs it; the next crawl will re-score it with the declaration taken into account.

## How this will be verified
Nothing is "verified safe" by assertion. When a new version is published, KillerSkills re-runs it in the honeypot and the verdict page updates; you can trigger it early from the verdict page ("Re-scan after I publish a fix"). Reply here or at [email protected] if you believe the verdict is wrong; disputes are reviewed by a person and the outcome is published on the verdict page.

_Proposed through killerskills.ai/threats by a user of the package, under their own GitHub account. The evidence is public; the package was never modified by KillerSkills._