Skill scan · first seen 2026-10-19 · scanned 1 time
detecting-ai-model-prompt-injection-attacks · mukul975/Anthropic-Cybersecurity-Skills
Skill scan · rules 2026-10-19.3
Medium-risk pattern
Caution: 1 pattern worth a look before installing.
| Severity | Pattern | Line | What it says |
|---|---|---|---|
| Medium | An instruction to ignore earlier instructions FS-INJ-001 The classic prompt-injection opening. | 87 | Ignore all previous instructions |
| Low | An instruction to ignore earlier instructions FS-INJ-001 The classic prompt-injection opening. | 146 | ignore previous instructions |
| Low | An instruction to ignore earlier instructions FS-INJ-001 The classic prompt-injection opening. | 157 | override the system prompt |
Fingerprintc6be759524e3…
sha256:c6be759524e385194b85e79b7d38545a9914c79d95c65ad946a2e0444527cdcd
Maintain this repository? Verify with GitHub to reply to these findings or report a fix. A person reviews every reply and aims to answer within 7 days.
Share
Checked a skill before my agent installed it. It has an instruction to ignore earlier instructions, so I read those lines first: https://theagentforge.io/scan/c6be759524e385194b85e79b7d38545a9914c79d95c65ad946a2e0444527cdcd?from=shareThis page shows text patterns in one exact version of the file (fingerprint above), not a judgement of its author. Publisher? Reply or dispute, and we add your reply here.
Check your own skills
Or paste this into your agent once. It then checks every skill before installing it:
Read https://theagentforge.io/skill.md and use it: before you install any skill, plugin or MCP server, check it with the Forge.Free, no account. Read the skill file first.