{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"code-and-sorts-promptdrifter","slug":"code-and-sorts-promptdrifter","name":"PromptDrifter","type":"cli","url":"https://github.com/Code-and-Sorts/PromptDrifter","page_url":"https://unfragile.ai/code-and-sorts-promptdrifter","categories":["testing-quality"],"tags":["anthropic-claude","ci-cd","command-line-tool","google-gemini","llm","ollama","openai","prompt-drift","prompt-engineering","prompt-tuning","python","regression-testing","testing","testing-tools"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"code-and-sorts-promptdrifter__cap_1","uri":"capability://safety.moderation.prompt.drift.detection.during.ci.builds","name":"prompt drift detection during ci builds","description":"PromptDrifter monitors the output of LLMs during CI builds by comparing current responses against previously recorded outputs. It employs a regression testing approach that flags any discrepancies, ensuring that any changes in LLM behavior are caught before deployment.","intents":["how to catch LLM output changes in CI","automatically validate LLM responses","ensure consistent LLM behavior in production"],"best_for":["developers deploying LLMs in production environments"],"limitations":["only detects changes in responses, not content quality"],"requires":["Python 3.7+","configured CI/CD pipeline"],"input_types":["text prompts","previous LLM responses"],"output_types":["detection alerts","build failure notifications"],"categories":["safety-moderation","testing-tools"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"code-and-sorts-promptdrifter__cap_2","uri":"capability://automation.workflow.one.command.integration.for.ci.cd","name":"one-command integration for ci/cd","description":"PromptDrifter allows users to set up prompt drift detection with a single command, simplifying the integration process into CI/CD workflows. This approach minimizes the overhead typically associated with configuring testing frameworks, enabling teams to focus on development rather than setup.","intents":["how to quickly integrate testing into CI/CD","simplify CI/CD setup for LLMs","fast track prompt testing in builds"],"best_for":["teams looking for quick CI/CD solutions"],"limitations":["may not support all CI/CD platforms out of the box"],"requires":["Python 3.7+","CI/CD toolchain access"],"input_types":["command-line inputs"],"output_types":["success messages","error reports"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"code-and-sorts-promptdrifter__headline","uri":"capability://automation.workflow.ci.guardrail.for.prompt.drift.detection","name":"ci guardrail for prompt drift detection","description":"PromptDrifter is a command-line tool designed to integrate into CI/CD pipelines, automatically detecting and failing builds when the responses from LLMs change unexpectedly. This ensures that prompt stability is maintained throughout the development lifecycle, providing developers with confidence in their LLM integrations.","intents":["best CI tool for LLM prompt testing","how to prevent prompt drift in CI/CD","LLM regression testing solutions","automated prompt validation for builds"],"best_for":["teams implementing CI/CD for LLM applications"],"limitations":["requires integration with existing CI/CD tools","only supports specific LLMs like OpenAI and Anthropic"],"requires":["Python 3.7+","access to CI/CD environment"],"input_types":["text prompts","LLM responses"],"output_types":["build status","error logs"],"categories":["automation-workflow","testing-tools"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":34,"verified":false,"data_access_risk":"moderate","permissions":["Python 3.7+","configured CI/CD pipeline","CI/CD toolchain access","access to CI/CD environment"],"failure_modes":["only detects changes in responses, not content quality","may not support all CI/CD platforms out of the box","requires integration with existing CI/CD tools","only supports specific LLMs like OpenAI and Anthropic","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0,"quality":0.41,"ecosystem":0.6000000000000001,"match_graph":0.25,"freshness":0.9,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.1,"match_graph":0.28,"freshness":0.12}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-28T10:40:41.893Z","last_scraped_at":null,"last_commit":"2026-05-27T11:03:09Z"},"community":{"stars":9,"forks":null,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=code-and-sorts-promptdrifter","compare_url":"https://unfragile.ai/compare?artifact=code-and-sorts-promptdrifter"}},"signature":"b7PiYJs8Cngwxe69t7MytxdU66LWoIJdZf8CB7yi9ZSr7TJNHrvZlXDidaEAVxISpe6wFjTOTjeu1K6jseNjDw==","signedAt":"2026-06-15T06:51:51.898Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/code-and-sorts-promptdrifter","artifact":"https://unfragile.ai/code-and-sorts-promptdrifter","verify":"https://unfragile.ai/api/v1/verify?slug=code-and-sorts-promptdrifter","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}