benchmarking-prompt-injection
Skill277 starsChanged 8 days agoAGPL-3.0View on GitHub
Use when changing or reviewing the prompt-injection detector's prompts or model selection, or when explicitly asked to run, investigate, or maintain its risk:pi benchmark, results, or fixtures. Do not activate for hook or upload handling, attachment ingestion, policy enforcement, evidence assembly, or other changes that leave detector prompts and models unchanged unless the task explicitly concerns the benchmark.
In this repository39 other files it gives its agents.
- .agents/skills/clickhouse-architecture-advisor/AGENTS.mdAGENTS.md
- .agents/skills/clickhouse-best-practices/AGENTS.mdAGENTS.md
- .agents/skills/vercel-react-best-practices/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- admin-shadcnSkill
- authoring-platform-mcp-skillsSkill
- clickhouse-architecture-advisorSkill
- clickhouse-best-practicesSkill
- clickhouseSkill
- datadogSkill
- feature-flagSkill
- frontendSkill
- glintSkill
- golangSkill
- gram-audit-loggingSkill
- gram-demo-seedSkill
- gram-functionsSkill
- gram-management-apiSkill
- gram-playwright-cliSkill
- gram-pubsub-pythonSkill
- gram-pubsubSkill
- gram-rbacSkill
- gram-telemetry-query-dimensionsSkill
- gram-temporalSkill
- implementing-mcp-provider-setupSkill
- maintaining-admin-mcpSkill
- maintaining-command-paletteSkill
- maintaining-platform-mcpSkill
- mise-tasksSkill
- page-toolbarSkill
- pitchforkSkill
- postgresqlSkill
- pull-request-demoSkill
- pull-requestSkill
- testing-local-slackSkill
- transactional-emailSkill
- vercel-react-best-practicesSkill
- writing-skillsSkill
- datadog-insightsSkill
The file
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.