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. 70 · Caution

Evidence

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

Diff · removes 5 caught lines

Download .patch

diff --git a/.claude/skills/claude-md-retro/SKILL.md b/.claude/skills/claude-md-retro/SKILL.md
--- a/.claude/skills/claude-md-retro/SKILL.md
+++ b/.claude/skills/claude-md-retro/SKILL.md
@@ -3,1 +3,0 @@
-description: Review this session's trace (tool calls, errors, retries, user corrections) and turn anything that failed or had to be reworked into a guardrail in CLAUDE.md, a path-scoped rule or a skil
@@ -8,1 +7,0 @@
-`/update-claude-md` keeps `CLAUDE.md` true to the **code**. This skill keeps the agent guidance effective based on the **process**: what went wrong in this session and what would have stopped it.
@@ -63,1 +61,0 @@
-- If you created, renamed or removed a skill or rule, update the "Keeping docs current" section of `CLAUDE.md`.
diff --git a/.claude/skills/update-claude-md/SKILL.md b/.claude/skills/update-claude-md/SKILL.md
--- a/.claude/skills/update-claude-md/SKILL.md
+++ b/.claude/skills/update-claude-md/SKILL.md
@@ -3,1 +3,0 @@
-description: Bring CLAUDE.md in line with the current team-memory codebase. Use after changes to architecture, module layout, commands, Settings fields, infra resources, conventions or gotchas, or whe
@@ -6,1 +5,0 @@
-# Update CLAUDE.md

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.1.0. 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: aws-team-memory: Instructs the agent to modify its persistent memory or instruction files (KillerSkills verdict caution 70)

## 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 70/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/glama/aws-team-memory/0.1.0

## Findings with a location
- `.claude/skills/claude-md-retro/SKILL.md:63` · Instructs the agent to modify its persistent memory or instruction files — .claude/skills/claude-md-retro/SKILL.md:63 · Instructs the agent to modify its persistent memory or instruction files
- `.claude/skills/claude-md-retro/SKILL.md:8` · Instructs the agent to modify its persistent memory or instruction files — .claude/skills/claude-md-retro/SKILL.md:8 · Instructs the agent to modify its persistent memory or instruction files
- `.claude/skills/update-claude-md/SKILL.md:6` · Instructs the agent to modify its persistent memory or instruction files — .claude/skills/update-claude-md/SKILL.md:6 · Instructs the agent to modify its persistent memory or instruction files
- `.claude/skills/claude-md-retro/SKILL.md:3` · Instructs the agent to modify its persistent memory or instruction files — .claude/skills/claude-md-retro/SKILL.md:3 · Instructs the agent to modify its persistent memory or instruction files
- `.claude/skills/update-claude-md/SKILL.md:3` · Instructs the agent to modify its persistent memory or instruction files — .claude/skills/update-claude-md/SKILL.md:3 · Instructs the agent to modify its persistent memory or instruction files

## 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 (pgstevenson/aws-team-memory)Fork to open a PR

# or from a terminal, under your own account:
gh repo fork pgstevenson/aws-team-memory --clone && cd aws-team-memory
curl -s https://killerskills.ai/skills/glama/aws-team-memory/0.1.0/patch.diff | git apply --unidiff-zero
git checkout -b killerskills/0.1.0 && git commit -am "aws-team-memory: Instructs the agent to modify its persistent memory or instruction files"
gh pr create --fill --body-file <(curl -s https://killerskills.ai/skills/glama/aws-team-memory/0.1.0/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.