Propose a patch
A fix the maintainer can act on, with the evidence attached. It opens under your GitHub account in your browser; KillerSkills holds no token and never modifies the package. The fix is verified the only way we verify anything: the next crawl re-runs the new version in the honeypot. 68 · Caution
Evidence
The sandbox observed nothing beyond the static findings; the proposal cites those.
Diff · removes 11 caught lines
Download .patchdiff --git a/sagrada_mcp/_vendor/sagrada_preflight/conflict.py b/sagrada_mcp/_vendor/sagrada_preflight/conflict.py
--- a/sagrada_mcp/_vendor/sagrada_preflight/conflict.py
+++ b/sagrada_mcp/_vendor/sagrada_preflight/conflict.py
@@ -245,1 +245,0 @@
-`https://user:sk-live-…@prod.example.com` put a live credential in a signed artifact,
@@ -838,1 +837,0 @@
-# A resource that is a stringified STRUCTURE — `'{"url": "https://u:SECRET@h"}'`,
@@ -850,1 +848,0 @@
-# format characters, so `{"url": "https://u:SECRET@h"}` kept a `` first
diff --git a/sagrada_mcp/_vendor/sagrada_tracker/battery.py b/sagrada_mcp/_vendor/sagrada_tracker/battery.py
--- a/sagrada_mcp/_vendor/sagrada_tracker/battery.py
+++ b/sagrada_mcp/_vendor/sagrada_tracker/battery.py
@@ -105,1 +105,0 @@
-"ByoLLMBattery needs a key — pass api_key=, or set SAGRADA_LLM_API_KEY "
diff --git a/tests/_fuzz_gate_invariants.py b/tests/_fuzz_gate_invariants.py
--- a/tests/_fuzz_gate_invariants.py
+++ b/tests/_fuzz_gate_invariants.py
@@ -89,1 +89,0 @@
-"https://u:tok@h/p?a=b", "/etc/hosts", "deploy://prod", "",
diff --git a/sagrada_mcp/_vendor/sagrada_preflight/gate.py b/sagrada_mcp/_vendor/sagrada_preflight/gate.py
--- a/sagrada_mcp/_vendor/sagrada_preflight/gate.py
+++ b/sagrada_mcp/_vendor/sagrada_preflight/gate.py
@@ -797,1 +797,0 @@
-the trailing space, matching no rule anyone would write), and CLAUDE.md truncated to
@@ -1821,1 +1820,0 @@
-Verified repro: `Write claude.md` returned a signed ALLOW and PHYSICALLY OVERWROTE
diff --git a/tests/test_mcp_surface_defects.py b/tests/test_mcp_surface_defects.py
--- a/tests/test_mcp_surface_defects.py
+++ b/tests/test_mcp_surface_defects.py
@@ -15,1 +15,0 @@
-measured ALLOW/HALT split on the same repo. `Write{file_path: "CLAUDE.md"}` — a
@@ -163,1 +162,0 @@
-("Write", {"file_path": "CLAUDE.md", "content": "x"}), # structured + relative
@@ -194,1 +192,0 @@
-"cd /tmp && echo hi > CLAUDE.md", # cwd genuinely moved before the write
diff --git a/tests/test_cmd_family_coverage.py b/tests/test_cmd_family_coverage.py
--- a/tests/test_cmd_family_coverage.py
+++ b/tests/test_cmd_family_coverage.py
@@ -304,1 +304,0 @@
-("ruby-dash-e", """ruby -e 'File.write("CLAUDE.md","x")'"""),
Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.3.1. 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: sagrada-mcp: URL with embedded credentials (KillerSkills verdict caution 68)
## 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 68/100**. Summary: no network, touches nothing outside the project, writes no memory. - See the findings below. Full trace and findings: https://killerskills.ai/skills/pypi/sagrada-mcp/0.3.1 ## Findings with a location - `sagrada_mcp/_vendor/sagrada_preflight/conflict.py:245` · URL with embedded credentials — sagrada_mcp/_vendor/sagrada_preflight/conflict.py:245 · URL with embedded credentials — may be phishing or credential leaking - `sagrada_mcp/_vendor/sagrada_preflight/conflict.py:850` · Hidden zero-width characters detected — sagrada_mcp/_vendor/sagrada_preflight/conflict.py:850 · Hidden zero-width characters detected — may conceal content - `sagrada_mcp/_vendor/sagrada_preflight/conflict.py:850` · URL with embedded credentials — sagrada_mcp/_vendor/sagrada_preflight/conflict.py:850 · URL with embedded credentials — may be phishing or credential leaking - `sagrada_mcp/_vendor/sagrada_preflight/conflict.py:838` · URL with embedded credentials — sagrada_mcp/_vendor/sagrada_preflight/conflict.py:838 · URL with embedded credentials — may be phishing or credential leaking - `sagrada_mcp/_vendor/sagrada_tracker/battery.py:105` · Credential pass-through configuration — sagrada_mcp/_vendor/sagrada_tracker/battery.py:105 · Credential pass-through configuration — agent relays sensitive credentials - `tests/_fuzz_gate_invariants.py:89` · URL with embedded credentials — tests/_fuzz_gate_invariants.py:89 · URL with embedded credentials — may be phishing or credential leaking - `sagrada_mcp/_vendor/sagrada_preflight/gate.py:1821` · Instructs the agent to modify its persistent memory or instruction files — sagrada_mcp/_vendor/sagrada_preflight/gate.py:1821 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `sagrada_mcp/_vendor/sagrada_preflight/gate.py:797` · Instructs the agent to modify its persistent memory or instruction files — sagrada_mcp/_vendor/sagrada_preflight/gate.py:797 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `tests/test_mcp_surface_defects.py:194` · Instructs the agent to modify its persistent memory or instruction files — tests/test_mcp_surface_defects.py:194 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `tests/test_cmd_family_coverage.py:304` · Instructs the agent to modify its persistent memory or instruction files — tests/test_cmd_family_coverage.py:304 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `tests/test_mcp_surface_defects.py:15` · Instructs the agent to modify its persistent memory or instruction files — tests/test_mcp_surface_defects.py:15 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `tests/test_mcp_surface_defects.py:163` · Instructs the agent to modify its persistent memory or instruction files — tests/test_mcp_surface_defects.py:163 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file ## Proposed change The attached patch removes the 11 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._
# or from a terminal, under your own account: gh repo fork Cruxia-Labs/sagrada-mcp --clone && cd sagrada-mcp curl -s https://killerskills.ai/skills/pypi/sagrada-mcp/0.3.1/patch.diff | git apply --unidiff-zero git checkout -b killerskills/0.3.1 && git commit -am "sagrada-mcp: URL with embedded credentials" gh pr create --fill --body-file <(curl -s https://killerskills.ai/skills/pypi/sagrada-mcp/0.3.1/patch.md)
We count proposals opened, not merged: a merge shows up as a new version that the crawl re-runs. Disputes: the maintainer can contest the verdict from its page; a person reviews it and the outcome is published.