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. 71 · Caution

Evidence

Read ~/work/acme-api/.npmrc, ~/.npmrc in the decoy home directory.

Diff · removes 5 caught lines

Download .patch

diff --git a/references/panelists.md b/references/panelists.md
--- a/references/panelists.md
+++ b/references/panelists.md
@@ -12,1 +12,0 @@
-- Command: `no_proxy='*' claude -p '<prompt>' --max-turns 1`
@@ -26,1 +25,0 @@
-- Command: `no_proxy='*' codex exec --skip-git-repo-check -c model_reasoning_effort="medium" '<prompt>'`
@@ -38,1 +36,0 @@
-- Command: `no_proxy='*' GEMINI_CLI_TRUST_WORKSPACE=true zsh -i -c 'gemini -p "<prompt>"'`
@@ -51,1 +48,0 @@
-- Command: `no_proxy='*' qwen -p '<prompt>'`
@@ -60,1 +56,0 @@
-- Command: `deepcode-panelist '<prompt>'`

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.6.7. 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: conclave-skill: Read ~/work/acme-api/.npmrc, ~/.npmrc in the decoy home directory (KillerSkills verdict caution 71)

## 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 71/100**. Summary: reads credential files.

- Read ~/work/acme-api/.npmrc, ~/.npmrc in the decoy home directory.

Full trace and findings: https://killerskills.ai/skills/clawhub/conclave-skill/1.6.7

## Findings with a location
- `references/panelists.md:12` · Fake instruction boundary tag — references/panelists.md:12 · Fake instruction boundary tag · informational: not an instruction file
- `references/panelists.md:51` · Fake instruction boundary tag — references/panelists.md:51 · Fake instruction boundary tag · informational: not an instruction file
- `references/panelists.md:38` · Fake instruction boundary tag — references/panelists.md:38 · Fake instruction boundary tag · informational: not an instruction file
- `references/panelists.md:26` · Fake instruction boundary tag — references/panelists.md:26 · Fake instruction boundary tag · informational: not an instruction file
- `references/panelists.md:60` · Fake instruction boundary tag — references/panelists.md:60 · Fake instruction boundary tag · informational: not an instruction file

## Proposed change
The attached patch removes the 5 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._

Open the issue on GitHub (MCLYang/conclave-skill)Fork to open a PR

# or from a terminal, under your own account:
gh repo fork MCLYang/conclave-skill --clone && cd conclave-skill
curl -s https://killerskills.ai/skills/clawhub/conclave-skill/1.6.7/patch.diff | git apply --unidiff-zero
git checkout -b killerskills/1.6.7 && git commit -am "conclave-skill: Read ~/work/acme-api/.npmrc, ~/.npmrc in the decoy home directory"
gh pr create --fill --body-file <(curl -s https://killerskills.ai/skills/clawhub/conclave-skill/1.6.7/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.