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. 52 · Caution
Evidence
The sandbox observed nothing beyond the static findings; the proposal cites those.
Diff · removes 8 caught lines
Download .patchdiff --git a/AGENTS.md b/AGENTS.md --- a/AGENTS.md +++ b/AGENTS.md @@ -33,1 +33,0 @@ -CRITICAL: This section is the most important part of this file. Vercel's evals showed passive context (always in prompt) achieves 100% pass rate vs 53% when age @@ -93,1 +92,0 @@ -- Run `agent-context promote --autopromote` to automatically append flagged patterns to AGENTS.md diff --git a/README.md b/README.md --- a/README.md +++ b/README.md @@ -97,1 +97,0 @@ -4. **Promote**: Run `agent-context promote` to review, or `--autopromote` to auto-append to `AGENTS.md` @@ -101,1 +100,0 @@ -<summary><strong>The Research (Why this works)</strong></summary> ### Agents have an instruction budget (~150-200 instructions) HumanLayer found frontier LLMs r @@ -180,1 +178,0 @@ -1. **Edit `AGENTS.md`** — Fill in your project name, stack, commands. Replace placeholders with real patterns from your codebase. diff --git a/SKILL.md b/SKILL.md --- a/SKILL.md +++ b/SKILL.md @@ -45,1 +45,0 @@ -Then edit `AGENTS.md` with your project specifics: name, stack, commands, actual patterns and gotchas from your codebase. This is the highest-leverage edit you'll make. @@ -278,1 +277,0 @@ -1. **Edit `AGENTS.md`.** Fill in your project name, stack, commands. Replace the placeholder patterns and gotchas with real ones from your codebase. This is the highest-leverage edit you'll make. diff --git a/skill-card.md b/skill-card.md --- a/skill-card.md +++ b/skill-card.md @@ -42,1 +42,0 @@ -- [GitHub: How to Write a Great AGENTS.md](https://github.blog/ai-and-ml/github-copilot/how-to-write-a-great-agents-md-lessons-from-over-2500-repositories/)
Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.3.2. 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: agent-context-system: HTML comment carrying imperative instructions to the agent (KillerSkills verdict caution 52)
## What the honeypot observed KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.9.2) and scored it **CAUTION 52/100**. Summary: static analysis only — behavior not yet observed. - See the findings below. Full trace and findings: https://killerskills.ai/skills/clawhub/agent-context-system/1.3.2 ## Findings with a location - `AGENTS.md:33` · HTML comment carrying imperative instructions to the agent — AGENTS.md:33 · HTML comment carrying imperative instructions to the agent - `AGENTS.md:93` · Instructs the agent to modify its persistent memory or instruction files — AGENTS.md:93 · Instructs the agent to modify its persistent memory or instruction files - `README.md:180` · Instructs the agent to modify its persistent memory or instruction files — README.md:180 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `SKILL.md:278` · Instructs the agent to modify its persistent memory or instruction files — SKILL.md:278 · Instructs the agent to modify its persistent memory or instruction files - `README.md:97` · Instructs the agent to modify its persistent memory or instruction files — README.md:97 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `skill-card.md:42` · Instructs the agent to modify its persistent memory or instruction files — skill-card.md:42 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `SKILL.md:45` · Instructs the agent to modify its persistent memory or instruction files — SKILL.md:45 · Instructs the agent to modify its persistent memory or instruction files - `README.md:101` · Collapsed <details> block carrying instructions — README.md:101 · Collapsed <details> block carrying instructions ## Proposed change The attached patch removes the 8 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 AndreaGriffiths11/agent-context-system --clone && cd agent-context-system curl -s https://killerskills.ai/skills/clawhub/agent-context-system/1.3.2/patch.diff | git apply --unidiff-zero git checkout -b killerskills/1.3.2 && git commit -am "agent-context-system: HTML comment carrying imperative instructions to the agent" gh pr create --fill --body-file <(curl -s https://killerskills.ai/skills/clawhub/agent-context-system/1.3.2/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.