REPOGEO REPORT · LITE
microsoftarchive/promptbench
Default branch main · commit fcda538b · scanned 6/27/2026, 12:26:55 PM
GitHub: 2,811 stars · 220 forks
Score trend below includes all ready runs (older left, newer right; scroll horizontally if needed). The table is collapsed by default—expand for newest-first rows, 10 per page.
3 ready scans. Expand the table below for newest-first rows (10 per page, paginated).
Action plan is what to do next — copy-pasteable changes prioritized by impact. Category visibility is the real GEO test: when a user asks an AI a brand-free question that should surface microsoftarchive/promptbench, does the AI actually recommend you — or your competitors? Objective checks verify the metadata signals AI engines weight first. Self-mention check detects whether AI even knows you exist by name.
Action plan — copy-paste fixes
3 prioritized changes generated by gemini-2.5-flash. Mark items done after you ship the fix.
- highabout#1Reposition the 'About' description to clarify its nature as an open-source library
Why:
CURRENTA unified evaluation framework for large language models
COPY-PASTE FIXAn open-source Python library and unified evaluation framework for large language models, designed for researchers and developers.
- mediumreadme#2Add a statement clarifying active maintenance despite 'microsoftarchive' designation
Why:
COPY-PASTE FIXAdd a prominent note near the top of the README, for example: 'Note on Repository Status: Despite the 'microsoftarchive' designation, this repository is actively maintained with regular updates and contributions. Please refer to the 'News and Updates' section for recent developments.'
- lowtopics#3Add more specific topics to reinforce its identity as a code library/framework
Why:
CURRENTadversarial-attacks, benchmark, chatgpt, evaluation, large-language-models, prompt, prompt-engineering, robustness
COPY-PASTE FIXadversarial-attacks, benchmark, chatgpt, evaluation, large-language-models, prompt, prompt-engineering, robustness, llm-evaluation-framework, python-library, open-source
Category GEO backends resolved for this scan: google/gemini-2.5-flash, deepseek/deepseek-v4-flash
Category visibility — the real GEO test
Brand-free queries asked to google/gemini-2.5-flash. Did AI recommend you, or someone else?
Same questions for every model — switch tabs to compare answers and rankings.
- Weights & Biases · recommended 1×
- Arize AI (Phoenix) · recommended 1×
- LangChain (LangSmith) · recommended 1×
- Galileo (LLM Studio) · recommended 1×
- Humanloop · recommended 1×
- CATEGORY QUERYSeeking a unified platform to evaluate and understand large language model behavior.you: not recommendedAI recommended (in order):
- Weights & Biases
- Arize AI (Phoenix)
- LangChain (LangSmith)
- Galileo (LLM Studio)
- Humanloop
- MLflow
- DeepEval
AI recommended 7 alternatives but never named microsoftarchive/promptbench. This is the gap to close.
Show full AI answer
- CATEGORY QUERYHow to assess the robustness of large language models against adversarial prompt attacks?you: #3AI recommended (in order):
- Garak (llm-attacks/garak)
- Adversarial GLUE (AdvGLUE)
- PromptBench (microsoft/promptbench) ← you
- TextAttack (textattack/textattack)
- Robustness Gym (RobustnessGym/RobustnessGym)
- OpenAI Evals (openai/evals)
- DeepMind's "Safety Gym"
Show full AI answer
Objective checks
Rule-based audits of metadata signals AI engines weight most.
- Metadata completenesspass
- README presencepass
Self-mention check
Does AI even know your repo exists when asked about it directly?
- Compared to common alternatives in this category, what is the core differentiator of microsoftarchive/promptbench?passAI named microsoftarchive/promptbench explicitly
AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?
- If a team adopts microsoftarchive/promptbench in production, what risks or prerequisites should they evaluate first?passAI did not name microsoftarchive/promptbench — likely talking about a different project
AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?
- In one sentence, what problem does the repo microsoftarchive/promptbench solve, and who is the primary audience?passAI named microsoftarchive/promptbench explicitly
AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?
Embed your GEO score
Drop this badge into the README of microsoftarchive/promptbench. It auto-updates whenever the report is rescanned and links back to the latest report — easy public proof that you care about AI discoverability.
[](https://repogeo.com/en/r/microsoftarchive/promptbench)<a href="https://repogeo.com/en/r/microsoftarchive/promptbench"><img src="https://repogeo.com/badge/microsoftarchive/promptbench.svg" alt="RepoGEO" /></a>Subscribe to Pro for deep diagnoses
microsoftarchive/promptbench — Lite scans stay free; this card itemizes Pro deep limits vs Lite.
- Deep reports10 / month
- Brand-free category queries5 vs 2 in Lite
- Prioritized action items8 vs 3 in Lite