Skip to content

aiskillstore/marketplace / skills/aayushbaniya2006/replicate-handler

replicate-handler

Verified

Integrate with Replicate AI for running models (image generation, LLMs, etc.).

FileKindSize
SKILL.mdskill md2.0 KB
reference.mdmarkdown11.4 KB
skill-report.jsonconfig22.9 KB

skill-report.json

View on GitHub
{
  "schema_version": "2.0",
  "meta": {
    "generated_at": "2026-07-09T12:36:46.931Z",
    "slug": "aayushbaniya2006-replicate-handler",
    "source_url": "https://github.com/AayushBaniya2006/blinkfund/tree/main/.claude/skills/replicate-handler",
    "source_ref": "main",
    "model": "codex",
    "analysis_version": "3.0.0",
    "source_type": "community",
    "content_hash": "01dec9c246f428dbc0a6fd1700e24bc873581c68f9e59af43905c6952c34d983",
    "tree_hash": "feda6288761f8351346aa39c5748e1fec5d8ff9df9f2295ba441c70b2e3defd5",
    "provenance": {
      "model_requested": "gpt-5.5:high",
      "model_effective": "codex:gpt-5.5:high",
      "fallback_chain": [
        {
          "agent": "codex",
          "model": "gpt-5.5",
          "effort": "high",
          "outcome": "success"
        }
      ],
      "analysis_version": "3.0.0"
    }
  },
  "skill": {
    "name": "replicate-handler",
    "description": "Use Replicate AI models with TypeScript examples for quick runs, long jobs, and SDK type references.",
    "summary": "Guides Replicate SDK setup, model execution, polling, and TypeScript integration.",
    "icon": "📦",
    "version": "1.0.0",
    "author": "AayushBaniya2006",
    "license": "MIT",
    "tags": [
      "replicate",
      "ai-models",
      "typescript",
      "image-generation",
      "inngest"
    ],
    "supported_tools": [
      "claude",
      "codex",
      "claude-code"
    ],
    "risk_factors": [
      "external_commands",
      "network",
      "env_access"
    ],
    "category": "coding"
  },
  "security_audit": {
    "risk_level": "safe",
    "is_blocked": false,
    "safe_to_publish": true,
    "analysis_status": "ok",
    "summary": "The audited files are a Replicate SDK usage guide and a TypeScript type reference. The static command, network, and sensitive-file alerts are explained by Markdown formatting, TypeScript syntax, or intended SDK authentication. No evidence of malware, prompt injection, credential exfiltration, or system reconnaissance was found.",
    "remediation": [
      {
        "issue": "Credential handling guidance is minimal.",
        "suggestion": "Add a warning to keep REPLICATE_API_TOKEN out of source control and rotate exposed tokens.",
        "severity": "low"
      },
      {
        "issue": "Package installation is shown without version guidance.",
        "suggestion": "Recommend installing a pinned Replicate SDK version and reviewing dependency changes before use.",
        "severity": "low"
      },
      {
        "issue": "Long-running model polling can create cost or resource risks.",
        "suggestion": "Advise timeouts, cancellation paths, and usage limits for expensive predictions.",
        "severity": "low"
      }
    ],
    "risk_factor_evidence": [
      {
        "factor": "external_commands",
        "evidence": [
          {
            "file": "reference.md",
            "line_start": 204,
            "line_end": 388
          },
          {
            "file": "reference.md",
            "line_start": 388,
            "line_end": 388
          },
          {
            "file": "reference.md",
            "line_start": 416,
            "line_end": 416
          },
          {
            "file": "reference.md",
            "line_start": 804,
            "line_end": 903
          },
          {
            "file": "SKILL.md",
            "line_start": 10,
            "line_end": 10
          },
          {
            "file": "SKILL.md",
            "line_start": 11,
            "line_end": 11
          },
          {
            "file": "SKILL.md",
            "line_start": 16,
            "line_end": 16
          },
          {
            "file": "SKILL.md",
            "line_start": 18,
            "line_end": 34
          },
          {
            "file": "SKILL.md",
            "line_start": 34,
            "line_end": 37
          },
          {
            "file": "SKILL.md",
            "line_start": 37,
            "line_end": 39
          },
          {
            "file": "SKILL.md",
            "line_start": 39,
            "line_end": 71
          },
          {
            "file": "SKILL.md",
            "line_start": 71,
            "line_end": 77
          }
        ]
      },
      {
        "factor": "network",
        "evidence": [
          {
            "file": "reference.md",
            "line_start": 436,
            "line_end": 436
          }
        ]
      },
      {
        "factor": "env_access",
        "evidence": [
          {
            "file": "SKILL.md",
            "line_start": 22,
            "line_end": 22
          },
          {
            "file": "SKILL.md",
            "line_start": 22,
            "line_end": 22
          }
        ]
      }
    ],
    "static_findings": [
      {
        "id": "external_commands:reference.md:204:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference.md",
        "line_start": 204,
        "line_end": 388,
        "snippet": "output?: any; // TODO: this should be `unknown`"
      },
      {
        "id": "external_commands:reference.md:388:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference.md",
        "line_start": 388,
        "line_end": 388,
        "snippet": "identifier: `${string}/${string}` | `${string}/${string}:${string}`,"
      },
      {
        "id": "external_commands:reference.md:416:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference.md",
        "line_start": 416,
        "line_end": 416,
        "snippet": "identifier: `${string}/${string}` | `${string}/${string}:${string}`,"
      },
      {
        "id": "external_commands:reference.md:804:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "reference.md",
        "line_start": 804,
        "line_end": 903,
        "snippet": "destination: `${string}/${string}`;"
      },
      {
        "id": "network:reference.md:436:http-client-library",
        "category": "network",
        "severity": "low",
        "pattern": "HTTP client library",
        "file": "reference.md",
        "line_start": 436,
        "line_end": 436,
        "snippet": "request("
      },
      {
        "id": "blocker:reference.md:408:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "reference.md",
        "line_start": 408,
        "line_end": 410,
        "snippet": "progress?: (prediction: Prediction) => void"
      },
      {
        "id": "external_commands:SKILL.md:10:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 10,
        "line_end": 10,
        "snippet": "1.  **Install**: `npm install replicate`"
      },
      {
        "id": "external_commands:SKILL.md:11:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 11,
        "line_end": 11,
        "snippet": "2.  **Environment**: Add `REPLICATE_API_TOKEN` to `.env` (and `.env.local`)."
      },
      {
        "id": "external_commands:SKILL.md:16:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 16,
        "line_end": 16,
        "snippet": "For models that complete quickly (seconds), use `replicate.run`."
      },
      {
        "id": "external_commands:SKILL.md:18:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 18,
        "line_end": 34,
        "snippet": "```typescript"
      },
      {
        "id": "external_commands:SKILL.md:34:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 34,
        "line_end": 37,
        "snippet": "```"
      },
      {
        "id": "external_commands:SKILL.md:37:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 37,
        "line_end": 39,
        "snippet": "For tasks that might timeout (video generation, large models), use Inngest's `step.sleep` to poll fo"
      },
      {
        "id": "external_commands:SKILL.md:39:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 39,
        "line_end": 71,
        "snippet": "```typescript"
      },
      {
        "id": "external_commands:SKILL.md:71:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 71,
        "line_end": 77,
        "snippet": "throw new Error(`Replicate failed: ${result.error}`);"
      },
      {
        "id": "env_access:SKILL.md:22:environment-variable-access-dot-notation",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable access (dot notation)",
        "file": "SKILL.md",
        "line_start": 22,
        "line_end": 22,
        "snippet": "auth: process.env.REPLICATE_API_TOKEN,"
      },
      {
        "id": "env_access:SKILL.md:22:environment-variable-object",
        "category": "env_access",
        "severity": "low",
        "pattern": "Environment variable object",
        "file": "SKILL.md",
        "line_start": 22,
        "line_end": 22,
        "snippet": "auth: process.env.REPLICATE_API_TOKEN,"
      },
      {
        "id": "sensitive:SKILL.md:11:environment-file-access",
        "category": "sensitive",
        "severity": "high",
        "pattern": "Environment file access",
        "file": "SKILL.md",
        "line_start": 11,
        "line_end": 11,
        "snippet": "2.  **Environment**: Add `REPLICATE_API_TOKEN` to `.env` (and `.env.local`)."
      },
      {
        "id": "sensitive:SKILL.md:22:environment-file-access",
        "category": "sensitive",
        "severity": "high",
        "pattern": "Environment file access",
        "file": "SKILL.md",
        "line_start": 22,
        "line_end": 22,
        "snippet": "auth: process.env.REPLICATE_API_TOKEN,"
      },
      {
        "id": "sensitive:SKILL.md:11:environment-variant-files",
        "category": "sensitive",
        "severity": "high",
        "pattern": "Environment variant files",
        "file": "SKILL.md",
        "line_start": 11,
        "line_end": 11,
        "snippet": "2.  **Environment**: Add `REPLICATE_API_TOKEN` to `.env` (and `.env.local`)."
      }
    ],
    "finding_verdicts": [
      {
        "id": "external_commands:reference.md:204:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "Line 204 is a TypeScript property using any for prediction output. There is no shell command, backtick execution, or executable code path."
      },
      {
        "id": "external_commands:reference.md:388:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.97,
        "reason": "Line 388 is a TypeScript template literal type for Replicate model identifiers. It is static type syntax, not runtime command execution."
      },
      {
        "id": "external_commands:reference.md:416:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.97,
        "reason": "Line 416 is another TypeScript template literal type for stream identifiers. It does not invoke a shell or evaluate user input."
      },
      {
        "id": "external_commands:reference.md:804:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.97,
        "reason": "Line 804 declares a TypeScript template literal type for a training destination. It is type documentation, not command execution."
      },
      {
        "id": "network:reference.md:436:http-client-library",
        "verdict": "false_positive",
        "confidence": 0.88,
        "reason": "Line 436 documents a request method in Replicate SDK type definitions. It shows intended API capability but no hidden endpoint or data exfiltration."
      },
      {
        "id": "blocker:reference.md:408:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "Line 408 is a progress callback receiving a Prediction object. It does not inspect the host system or perform reconnaissance."
      },
      {
        "id": "external_commands:SKILL.md:10:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.78,
        "reason": "The line shows a hardcoded npm install command in Markdown setup instructions. It is not shell backtick execution or dynamic command construction."
      },
      {
        "id": "external_commands:SKILL.md:11:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.91,
        "reason": "The backticked values are an environment variable name and file names in documentation. They do not execute commands."
      },
      {
        "id": "external_commands:SKILL.md:16:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The line names the Replicate SDK method replicate.run in prose. It is not Ruby or shell backtick execution."
      },
      {
        "id": "external_commands:SKILL.md:18:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The finding points to a Markdown code fence for a TypeScript example. The example calls the Replicate SDK and contains no shell execution."
      },
      {
        "id": "external_commands:SKILL.md:34:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The finding points to the closing Markdown code fence. It is formatting only and has no executable behavior."
      },
      {
        "id": "external_commands:SKILL.md:37:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "The backticked text names Inngest step.sleep in prose. It describes polling behavior, not shell execution."
      },
      {
        "id": "external_commands:SKILL.md:39:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The finding points to a Markdown code fence for an Inngest TypeScript example. The code uses SDK calls and durable sleeps, not external commands."
      },
      {
        "id": "external_commands:SKILL.md:71:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "Line 71 uses a TypeScript template string to build an error message from Replicate status. It does not execute the string as a command."
      },
      {
        "id": "env_access:SKILL.md:22:environment-variable-access-dot-notation",
        "verdict": "false_positive",
        "confidence": 0.82,
        "reason": "The example reads REPLICATE_API_TOKEN only as SDK authentication. I found no evidence that the token is logged, exposed, or sent to an unrelated service."
      },
      {
        "id": "env_access:SKILL.md:22:environment-variable-object",
        "verdict": "false_positive",
        "confidence": 0.82,
        "reason": "The process.env usage is limited to the expected Replicate API token field. This is normal credential configuration for the SDK example."
      },
      {
        "id": "sensitive:SKILL.md:11:environment-file-access",
        "verdict": "false_positive",
        "confidence": 0.84,
        "reason": "The line tells users where to store a Replicate token, but it does not read .env files or disclose their contents. It is setup guidance, not file access."
      },
      {
        "id": "sensitive:SKILL.md:22:environment-file-access",
        "verdict": "false_positive",
        "confidence": 0.8,
        "reason": "The code sample references a single environment variable for intended API authentication. No file read, credential dump, or unrelated network transfer is present."
      },
      {
        "id": "sensitive:SKILL.md:11:environment-variant-files",
        "verdict": "false_positive",
        "confidence": 0.84,
        "reason": "The .env and .env.local references are documentation for local configuration. They are not instructions to inspect or exfiltrate environment files."
      }
    ],
    "semantic_findings": [],
    "critical_findings": [],
    "high_findings": [],
    "medium_findings": [],
    "low_findings": [],
    "dangerous_patterns": [],
    "files_scanned": 2,
    "total_lines": 986,
    "audit_model": "codex",
    "audited_at": "2026-07-09T12:36:46.931Z",
    "agent_auto_install_policy": "allowed",
    "manual_install_policy": "allowed"
  },
  "content": {
    "user_title": "Run Replicate Models in Apps",
    "value_statement": "Replicate integrations can be unclear when jobs finish at different speeds. This skill gives Claude, Codex, and Claude Code practical patterns for quick runs, polling, and typed responses.",
    "seo_keywords": [
      "Replicate AI",
      "Replicate SDK",
      "Claude",
      "Codex",
      "Claude Code",
      "AI model integration",
      "image generation",
      "TypeScript Replicate",
      "Inngest polling",
      "webhook validation"
    ],
    "actual_capabilities": [
      "Shows how to install the Replicate SDK and configure REPLICATE_API_TOKEN.",
      "Provides a quick run pattern for short Replicate model jobs.",
      "Provides an Inngest polling pattern for long-running predictions.",
      "Includes TypeScript references for predictions, streams, trainings, and webhooks.",
      "Shows failure handling when Replicate returns a failed prediction status."
    ],
    "limitations": [
      "Does not include a complete application or runnable project scaffold.",
      "Requires a valid Replicate account and API token.",
      "Does not choose model versions or manage model costs for users.",
      "Type references may lag behind newer Replicate SDK releases."
    ],
    "use_cases": [
      {
        "title": "Add Image Generation",
        "description": "Use the quick run pattern to call an image model from a TypeScript feature.",
        "target_user": "Application developer"
      },
      {
        "title": "Build Durable Video Jobs",
        "description": "Use the polling pattern when model output may take longer than one request.",
        "target_user": "Backend engineer"
      },
      {
        "title": "Type Replicate Responses",
        "description": "Use the reference file to understand prediction, webhook, and training shapes.",
        "target_user": "TypeScript maintainer"
      }
    ],
    "prompt_templates": [
      {
        "title": "Set Up Replicate",
        "scenario": "Beginner setup",
        "prompt": "Use the replicate-handler skill to explain the minimum setup for Replicate in my TypeScript app."
      },
      {
        "title": "Run A Short Model Job",
        "scenario": "Basic model execution",
        "prompt": "Use the replicate-handler skill to adapt a quick Replicate run pattern for a short image generation task."
      },
      {
        "title": "Design Long Job Polling",
        "scenario": "Durable background work",
        "prompt": "Use the replicate-handler skill to design an Inngest polling flow for a long Replicate video job."
      },
      {
        "title": "Review Types And Webhooks",
        "scenario": "Advanced integration review",
        "prompt": "Use the replicate-handler skill to review my Replicate integration plan for response types, webhooks, and failure handling."
      }
    ],
    "output_examples": [
      {
        "input": "Ask for a quick Replicate image generation pattern.",
        "output": "A concise plan covering SDK setup, token configuration, model selection, input preparation, and result handling."
      },
      {
        "input": "Ask how to handle a long-running Replicate video task.",
        "output": "A polling workflow using durable wait steps, status checks, failure handling, and final output delivery."
      },
      {
        "input": "Ask what Replicate response types to expect.",
        "output": "A plain-language summary of prediction status, output fields, errors, logs, webhooks, and training references."
      }
    ],
    "best_practices": [
      "Store REPLICATE_API_TOKEN outside source control and rotate it if exposed.",
      "Use quick runs only for jobs that complete within request time limits.",
      "Poll long-running predictions with durable steps and clear failure handling."
    ],
    "anti_patterns": [
      "Do not paste Replicate tokens into prompts, logs, or source files.",
      "Do not run expensive models without rate limits or user confirmation.",
      "Do not poll inside a busy loop or block server resources."
    ],
    "faq": [
      {
        "question": "What does this skill help with?",
        "answer": "It helps agents integrate the Replicate SDK, run models, poll predictions, and understand TypeScript response shapes."
      },
      {
        "question": "Does it run Replicate models by itself?",
        "answer": "No. It provides guidance and examples that an agent can adapt inside your project."
      },
      {
        "question": "Which API token does it expect?",
        "answer": "It expects a Replicate API token stored as REPLICATE_API_TOKEN in your environment."
      },
      {
        "question": "When should I use polling?",
        "answer": "Use polling for video generation, large models, or any prediction that may outlive one request."
      },
      {
        "question": "Does it include webhook guidance?",
        "answer": "The reference file includes webhook-related SDK types and validation function signatures."
      },
      {
        "question": "Can it help with Claude Code and Codex?",
        "answer": "Yes. The content is compatible with Claude, Codex, and Claude Code workflows."
      }
    ]
  },
  "file_structure": [
    {
      "name": "reference.md",
      "type": "file",
      "path": "reference.md",
      "lines": 905
    },
    {
      "name": "SKILL.md",
      "type": "file",
      "path": "SKILL.md",
      "lines": 81
    }
  ]
}
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall