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. 76 · Caution
Evidence
The sandbox observed nothing beyond the static findings; the proposal cites those.
Diff · removes 3 caught lines
Download .patchdiff --git a/src/lore/cli/commands/init.py b/src/lore/cli/commands/init.py
--- a/src/lore/cli/commands/init.py
+++ b/src/lore/cli/commands/init.py
@@ -140,1 +140,0 @@
-"""Add @MCP_Lore.md import to CLAUDE.md under MCP Documentation section.
@@ -358,1 +357,0 @@
-api_key = Prompt.ask(" Enter your API key (or press Enter to skip)", password=True)
@@ -377,1 +375,0 @@
-# Add import to CLAUDE.md (SuperClaude style)
Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.0.17. 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: lore-mcp: Prompting user to enter sensitive credentials (KillerSkills verdict caution 76)
## What the honeypot observed KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.7) and scored it **CAUTION 76/100**. Summary: no network, touches nothing outside the project, writes no memory. - See the findings below. Full trace and findings: https://killerskills.ai/skills/pypi/lore-mcp/1.0.17 ## Findings with a location - `src/lore/cli/commands/init.py:358` · Prompting user to enter sensitive credentials — src/lore/cli/commands/init.py:358 · Prompting user to enter sensitive credentials — social engineering - `src/lore/cli/commands/init.py:377` · Instructs the agent to modify its persistent memory or instruction files — src/lore/cli/commands/init.py:377 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `src/lore/cli/commands/init.py:140` · Instructs the agent to modify its persistent memory or instruction files — src/lore/cli/commands/init.py:140 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file ## Proposed change The attached patch removes the 3 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 ctx-stack/lore-mcp --clone && cd lore-mcp curl -s https://killerskills.ai/skills/pypi/lore-mcp/1.0.17/patch.diff | git apply --unidiff-zero git checkout -b killerskills/1.0.17 && git commit -am "lore-mcp: Prompting user to enter sensitive credentials" gh pr create --fill --body-file <(curl -s https://killerskills.ai/skills/pypi/lore-mcp/1.0.17/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.