CategoryWeightCategory scoreContribution

Securityprompt, exec, net, exfil, eval

35%

26.3 pts

Supply chainhash, typosquat, maintainer, lockfile

20%

100

20.0 pts

Maintenancestaleness, pinning, CI

15%

100

15.0 pts

TransparencySKILL.md, perms, README

15%

100

15.0 pts

Communityinstalls, verify, response

15%

100

15.0 pts

Findings & checks · 1 flagged

Securityscore 75 · 1 finding

HIGH"Ignore previous instructions" command embedded in the skillSS-SKILL-INJECT-IGNORE-01 · Prompt injection · CHANGELOG.md

HIGH — when it fires on hostile content the impact is full system-prompt override.

Why it matters

The text {match} is the classic direct prompt-injection phrasing. Placed in a skill body that the agent reads as trusted instructions, it tries to make the agent abandon its prior rules and follow whatever comes next — a full system-prompt override.

The exact value spotted

excerptCHANGELOG.md· markdown

9- 🗑️ **Removed** `chore(skills)`: **the `readme-generator` skill, superseded by the standal

… (109 chars elided on L9)

11- 🔒 **Added** `feat(security)`: **an Untrusted Content Boundary — instructions found inside

… (109 chars elided on L11)

13The classification lives in **one** place. A skill-audit table in the rule maps every skill

… (106 chars elided on L13)

Occurrences

1 occurrence · at L11

How to fix

Remove the override phrase, or rephrase the passage so it does not instruct the agent to discard its rules.

Delete the ignore/disregard/forget … previous instructions sentence.
If this is jailbreak-research or tutorial content, move the example into a clearly fenced, non-instruction block and label it as a quoted sample.

Framework references

OWASPLLM01 ATLASAML.T0051

Trace & refs

ruleSS-SKILL-INJECT-IGNORE-01sha2562e4221a7f996a729…rubric 365aacaView on GitHub

Supply chainscore 100 · 0 findings

✓—All supply chain checks passedNo findings in this category for the latest scan.pass

Maintenancescore 100 · 0 findings

✓—All maintenance checks passedNo findings in this category for the latest scan.pass

Transparencyscore 100 · 0 findings

✓—All transparency checks passedNo findings in this category for the latest scan.pass

Communityscore 100 · 0 findings

✓—All community checks passedNo findings in this category for the latest scan.pass

The primary manifest — the file an agent reads to learn what this artifact does.

skills/create-issue/SKILL.md · 1 file

skills/create-issue/SKILL.md1.1 KB · Markdown

name: create-issue

description: Use when create a single issue from provided text without modifying its content

license: MIT

metadata:

author: Petr Král (pekral.cz)

Create Issue

Purpose

Create a well-formatted issue while preserving the original content exactly.

Constraints

Preserve original text exactly (no rewriting or summarizing)
Improve formatting only (headings, lists, spacing)
Assign issue to current user
Use available CLI tools

Execution

1. Prepare Content

Use original text as-is
Apply readable formatting (headings, lists)

2. Generate Title

Use first line of input
Remove formatting only

3. Create Issue

Use available CLI tool
Set title and formatted description
Assign the most relevant existing label (per @rules/compound-engineering/general.mdc Assign the most relevant existing label when creating a tracker issue)
Assign to current user

4. Output

Return direct link to created issue

Output Format

# Task

<original text>

---

# Notes

Formatted automatically. Content unchanged.

create-issue— agent skill

Is create-issue safe to install?

Create Issue

Purpose

Constraints

Execution

1. Prepare Content

2. Generate Title

3. Create Issue

4. Output

Output Format

Audit the pieces. Scan the whole. Decide.