ai.plith/plith

Est-ce que ai.plith/plith est sûr ?

Au dernier audit MCP Hub, le serveur MCP « ai.plith/plith » est classé « Non encore testé ». Référencé dans l'annuaire. Seul le pentest comportemental fait foi, et il n'a pas encore été exécuté sur ce serveur. Le verdict est réévalué en continu ; consultez la fiche pour l'état à jour.

Méthode : MCP Hub analyse la description de chaque outil (couche 1) et, pour le verdict fort, exécute le serveur dans une micro-VM isolée avec leurres et réseau bloqué (couche 2). On publie des noms et des compteurs, jamais le contenu brut renvoyé par un outil.

Les 15 outils

• burnrate_budget≈ 310 tokens

  Get today's tracked LLM spend, per-model breakdown, projection, and budget alerts. Free — no credits charged.

  Paramètres, retour & exemple (1)

  Paramètres

  • daily_limit · number · optionnel
    Optional. Daily budget in USD (e.g., 10.0 for a $10/day cap). Enables budget alerts and remaining-balance calculation.

  Exemple d'appel

  {"name": "burnrate_budget", "arguments": {"daily_limit": 1}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"date": "texte", "spend": {}, "alerts": [], "projection": {}, "request_id": "texte", "credits_used": 1, "credits_remaining": 1, "fallback_behavior": "texte"}}

• burnrate_estimate≈ 557 tokens

  Before executing a multi-step agent plan, estimate the total LLM cost. Returns per-step breakdown and optimization suggestions. If the estimate exceeds your budget, pipe the same plan into burnrate_optimize. Costs 1 credit.

  Paramètres, retour & exemple (1)

  Paramètres

  • plan · array · requis
    Array of plan steps with provider, model, and token estimates.

  Exemple d'appel

  {"name": "burnrate_estimate", "arguments": {"plan": []}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"estimate": {}, "request_id": "texte", "credits_used": 1, "credits_remaining": 1, "fallback_behavior": "texte", "optimization_suggestions": []}}

• burnrate_optimize≈ 536 tokens

  Get a cheaper equivalent plan by substituting models with lower-cost alternatives. Call after burnrate_estimate if the estimated cost exceeds your budget. Returns the optimized plan with substituted models, new per-step costs, total savings, and whether the target_budget is met. Optionally set target_budget to constrain the optimization. Costs 1 credit.

  Paramètres, retour & exemple (2)

  Paramètres

  • plan · array · requis
    Array of plan steps. Same schema as burnrate_estimate: each step needs step, provider, model, estimated_input_tokens, estimated_output_tokens.
  • target_budget · number · optionnel
    Optional. Target total cost in USD.

  Exemple d'appel

  {"name": "burnrate_optimize", "arguments": {"plan": []}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"steps": [], "original": {}, "optimized": {}, "request_id": "texte", "suggestions": [], "credits_used": 1, "credits_remaining": 1, "fallback_behavior": "texte"}}

• burnrate_track≈ 530 tokens

  Log the actual cost of an LLM call after execution. Call this after every LLM request to build calibration data that improves burnrate_estimate accuracy over time. Free — no credits charged. Returns the recorded cost entry with computed margin versus the prior estimate when one exists for this model and token range.

  Paramètres, retour & exemple (6)

  Paramètres

  • model · string · requis
    Model identifier as returned by the provider. Examples: claude-sonnet-4-6, gpt-4o, gemini-2.0-flash, mistral-large-latest. Unknown models are accepted but cost may show as $0.
  • task_id · string · optionnel
    Optional task ID for cross-referencing spend with DedupQ deduplication results. Use the same task_id passed to dedupq_check to link cost tracking with deduplication.
  • provider · string · requis
    LLM provider identifier. Supported: anthropic, openai, google, mistral, cohere, deepseek, together, fireworks, groq. Must match the provider of the model used.
  • input_tokens · number · requis
    Actual prompt tokens used. Must be >= 0.
  • output_tokens · number · requis
    Actual completion tokens used. Must be >= 0.
  • cache_read_tokens · number · optionnel
    Optional. Cache-read tokens.

  Exemple d'appel

  {"name": "burnrate_track", "arguments": {"model": "texte", "provider": "texte", "input_tokens": 1, "output_tokens": 1}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"model": "texte", "tracked": true, "provider": "texte", "record_id": "texte", "request_id": "texte", "credits_used": 1, "input_tokens": 1, "output_tokens": 1, "pricing_found": true, "actual_cost_usd": 1, "credits_remaining": 1, "fallback_behavior": "texte"}}

• dedupq_check≈ 417 tokens

  Before executing any LLM task, check if an identical or semantically similar task has already been completed. Returns cached result on hit, saving one LLM call. On a miss, execute your task and call dedupq_complete to cache the result for future hits. Costs 1 credit.

  Paramètres, retour & exemple (4)

  Paramètres

  • content · string · requis
    The task content to check for duplicates. This is hashed and embedded for matching.
  • task_id · string · optionnel
    Optional caller task ID for tracing and cross-referencing with BurnRate.
  • hash_only · boolean · optionnel
    If true, skip vector similarity search and use exact hash matching only. Default: false.
  • similarity_threshold · number · optionnel
    Cosine similarity threshold for semantic matching, 0.0 to 1.0. Default: 0.80.

  Exemple d'appel

  {"name": "dedupq_check", "arguments": {"content": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"match": {}, "status": "texte", "cache_hit": "texte", "request_id": "texte", "content_hash": "texte", "credits_used": 1, "cache_age_seconds": 1, "credits_remaining": 1, "fallback_behavior": "texte"}}

• dedupq_complete≈ 302 tokens

  After executing a task, store the result so future identical or similar tasks return a cache hit via dedupq_check. Costs 2 credits.

  Paramètres, retour & exemple (4)

  Paramètres

  • result · ? · requis
    The task result to cache. Can be any JSON value.
  • content · string · requis
    Original task content. Used to compute hash and embedding for future matching.
  • task_id · string · optionnel
    Optional task ID. Used as the database row ID if provided.
  • hash_only · boolean · optionnel
    If true, skip embedding generation. Default: false.

  Exemple d'appel

  {"name": "dedupq_complete", "arguments": {"result": "texte", "content": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"stored": true, "task_id": "texte", "request_id": "texte", "content_hash": "texte", "credits_used": 1, "has_embedding": true, "credits_remaining": 1, "fallback_behavior": "texte"}}

• guardrail_check≈ 350 tokens

  Evaluate a proposed agent action against your governance policies. Returns allow or deny with the matched policy reason. Requires at least one active policy created via guardrail_create_policy. Deterministic rule evaluation — no LLM. Costs 1 credit.

  Paramètres, retour & exemple (2)

  Paramètres

  • agent_id · string · requis
    Agent identifier.
  • proposed_action · object · requis
    Action to evaluate. Must contain a 'type' field. Example: {"type": "http_request", "url": "https://external.example.com"} or {"type": "file_write", "path": "/etc/config"}.

  Exemple d'appel

  {"name": "guardrail_check", "arguments": {"agent_id": "texte", "proposed_action": {}}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"reason": "texte", "audit_id": "texte", "decision": "allow", "policy_id": "texte", "request_id": "texte", "credits_used": 1, "credits_remaining": 1, "fallback_behavior": "texte"}}

• guardrail_create_policy≈ 762 tokens

  Create a persistent governance policy that guardrail_check evaluates on every subsequent call. Define rules using and/or/not operators over action types, resource patterns, and budget thresholds. Call this before using guardrail_check — checks require at least one active policy. Policies persist until explicitly deleted. Duplicate policy names return an error. Returns the created policy with its ID and active status.

  Paramètres, retour & exemple (5)

  Paramètres

  • name · string · requis
    Unique policy name per org. Examples: 'no-delete-in-prod', 'budget-cap-50', 'pii-block'.
  • rules · array · requis
    Array of rule objects evaluated against the proposed_action in guardrail_check. Leaf operators: eq, starts_with, contains, gt, lt (compare field to value). Compound operators: and, or, not (nest sub-rules in a rules array). Example: [{operator:'eq', field:'type', value:'file_write'}] blocks all file writes. Nested example: [{operator:'and', rules:[{operator:'eq',field:'type',value:'api_call'},{operator:'contains',field:'url',value:'prod'}]}] blocks prod API calls.
  • priority · number · optionnel
    Optional. Evaluation order. Default: 0.
  • description · string · optionnel
    Optional human-readable summary of what this policy enforces. Returned in guardrail_check responses and guardrail_list_policies output for auditability.
  • action_types · array · optionnel
    Optional. Restrict this policy to only evaluate when proposed_action.type matches one of these values. Examples: ['file_write', 'api_call', 'db_delete']. Omit to apply the policy to all action types regardless of type field.

  Exemple d'appel

  {"name": "guardrail_create_policy", "arguments": {"name": "texte", "rules": []}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"policy": {}, "request_id": "texte", "credits_used": 1, "credits_remaining": 1, "fallback_behavior": "texte"}}

• pitfalldb_query≈ 421 tokens

  Check for known failure patterns before executing a task type. Returns pitfalls with severity, fix suggestions, and confidence scores. After your agent runs, submit failures via pitfalldb_report so others benefit. Costs 2 credits.

  Paramètres, retour & exemple (3)

  Paramètres

  • filters · object · optionnel
    Optional filters.
  • task_type · string · requis
    Task category: code_generation, web_search, data_analysis, etc.
  • task_description · string · optionnel
    Optional. Natural-language task description for semantic search.

  Exemple d'appel

  {"name": "pitfalldb_query", "arguments": {"task_type": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"pitfalls": [], "request_id": "texte", "credits_used": 1, "total_matching": 1, "credits_remaining": 1, "fallback_behavior": "texte"}}

• pitfalldb_report≈ 418 tokens

  Report an agent failure. PII-scrubbed before storage. Linked to existing pitfalls if similar. Free — no credits charged.

  Paramètres, retour & exemple (3)

  Paramètres

  • failure · object · requis
    Failure details.
  • task_type · string · requis
    Task category.
  • task_description · string · requis
    Description of the failed task.

  Exemple d'appel

  {"name": "pitfalldb_report", "arguments": {"failure": {}, "task_type": "texte", "task_description": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"message": "texte", "verified": true, "report_id": "texte", "request_id": "texte", "credits_used": 1, "credits_remaining": 1, "fallback_behavior": "texte", "linked_pitfall_id": "texte"}}

• qualitygate_validate≈ 662 tokens

  After your agent generates output, validate it against your rules before shipping. Runs deterministic checks (regex, JSON schema, syntax) plus optional LLM-powered tone and factual analysis. Returns a structured verdict (pass, warn, or fail) with a 0-100 score and per-check issue details. Use qualitygate_trends to spot recurring failure patterns over time. Variable cost: 1 credit per deterministic check, 8 credits per LLM check.

  Paramètres, retour & exemple (7)

  Paramètres

  • output · string · requis
    The agent output text to validate.
  • schema · object · optionnel
    JSON Schema to validate output against.
  • language · string · optionnel
    Code language for syntax check: json, python, javascript, typescript.
  • override · boolean · optionnel
    Force pass. Requires override_reason.
  • directives · array · optionnel
    Directive objects. Types: must_include, must_not_include, must_match, must_not_match, must_contain, must_not_contain, min_length, max_length.
  • check_types · array · optionnel
    Checks to run. Auto-inferred if omitted.
  • override_reason · string · optionnel
    Required when override is true.

  Exemple d'appel

  {"name": "qualitygate_validate", "arguments": {"output": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"issues": [], "summary": {}, "verdict": "pass", "checks_run": [], "request_id": "texte", "credits_used": 1, "credits_remaining": 1, "fallback_behavior": "texte"}}

• rigor_execute≈ 1951 tokens

  Execute a structured workflow end-to-end. Call rigor_plan first (free) to preview the step sequence and cost estimate before committing credits. Classifies the task, selects the optimal tool sequence, and executes each step with the right LLM model. Returns a complete deliverable — solution designs, competitive analyses, governance documents, and more. Supports SSE streaming for real-time progress, webhook callback, or polling. For atomic work — classification, scoring, ranking, entity extraction, query parsing — set preferences.execution to 'direct' and declare preferences.output_contract to get validated JSON records from a single call, routed to the cheapest model that holds the schema.

  Paramètres, retour & exemple (7)

  Paramètres

  • context · object · optionnel
    Additional context for the workflow.
  • delivery · object · optionnel
    Delivery method. Default: polling (MCP clients typically can't handle SSE).
  • task_type · string · optionnel
    Optional hint to bypass automatic classification. Passing it also removes the slowest classification tiers from the critical path, so send it whenever you know the shape. Multi-step deliverable types: solution_design, requirements_analysis, code_implementation, code_review, bug_fix, root_cause_analysis, incident_response, deployment_execution, competitive_scan, financial_analysis, research_task, documentation, governance_change, compliance_audit, data_security_assessment, performance_optimization, user_story_definition, implementation_prompt_generation. Atomic single-call types, which auto-select direct execution: tag, score, rerank, compose, extract_entities, parse_query, quick_research, quick_classification, quick_extraction, quick_scoring. Call GET /api/rigor/task-types for the full vocabulary with each type's shape.
  • preferences · object · optionnel
    Optional workflow preferences.
  • context_fields · array · optionnel
    Optional, with prior_workflow_id: the fields to carry over. Pass ["full_step_results"] for every step output in full.
  • task_description · string · requis
    Natural language description of the task, at most 4,000 characters. Be specific — include what you want produced, constraints, and context. Example: 'Design a caching layer for our API gateway with Redis integration.'
  • prior_workflow_id · string · optionnel
    Optional: the workflow_id (wr_…) of a finished workflow in your workspace to build on. Its results are passed to this run as context. It is charged as a new run at the normal quote.

  Exemple d'appel

  {"name": "rigor_execute", "arguments": {"task_description": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"ok": true, "status": "texte", "poll_url": "texte", "execution": "texte", "task_type": "texte", "value_class": "texte", "workflow_id": "texte", "delivery_mode": "texte", "available_modes": [], "estimated_credits": 1, "execution_fallback": true}}

• rigor_plan≈ 2022 tokens

  Before executing a complex task, get a structured workflow plan with per-step cost estimates. Classifies your task, selects the optimal framework sequence, and returns the full plan without executing anything. The response's allowed_modes tells you whether this plan is eligible for direct execution. Free — no credits charged.

  Paramètres, retour & exemple (5)

  Paramètres

  • task_type · string · optionnel
    Optional hint to bypass automatic classification. Passing it also removes the slowest classification tiers from the critical path, so send it whenever you know the shape. Multi-step deliverable types: solution_design, requirements_analysis, code_implementation, code_review, bug_fix, root_cause_analysis, incident_response, deployment_execution, competitive_scan, financial_analysis, research_task, documentation, governance_change, compliance_audit, data_security_assessment, performance_optimization, user_story_definition, implementation_prompt_generation. Atomic single-call types, which auto-select direct execution: tag, score, rerank, compose, extract_entities, parse_query, quick_research, quick_classification, quick_extraction, quick_scoring. Call GET /api/rigor/task-types for the full vocabulary with each type's shape.
  • preferences · object · optionnel
    Optional workflow preferences.
  • context_fields · array · optionnel
    Optional, with prior_workflow_id: the fields to carry over. Pass ["full_step_results"] for every step output in full.
  • task_description · string · requis
    Natural language description of the task, at most 4,000 characters. Be specific — include what you want produced, constraints, and context. Example: 'Design a caching layer for our API gateway with Redis integration.'
  • prior_workflow_id · string · optionnel
    Optional: the workflow_id (wr_…) of a finished workflow in your workspace to build on. Its results are passed to this run as context. It is charged as a new run at the normal quote.

  Exemple d'appel

  {"name": "rigor_plan", "arguments": {"task_description": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"ok": true, "plan": {}, "generated_title": "texte"}}

• rigor_status≈ 509 tokens

  Check the status of a running or completed Rigor workflow. Returns progress, step results, and the full deliverable when complete. Use after rigor_execute with polling delivery to retrieve results.

  Paramètres, retour & exemple (1)

  Paramètres

  • workflow_id · string · requis
    The workflow ID returned by rigor_execute (format: wr_xxx).

  Exemple d'appel

  {"name": "rigor_status", "arguments": {"workflow_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"ok": true, "workflow": {}}}

• rigor_workflows≈ 818 tokens

  List and search Rigor workflows for your organization, with filtering and pagination. Returns status, progress, capacity usage, and available actions per workflow. Use to monitor workflow state, understand concurrent limit usage, identify stuck or completed workflows, and — via q — find prior work on a subject before commissioning it again. Pair a q hit with rigor_status to read that workflow's deliverable.

  Paramètres, retour & exemple (9)

  Paramètres

  • q · string · optionnel
    Search the workflow title and task description. Every whitespace-separated term must appear in one or the other, as a case-insensitive substring — so "vector search postgres" matches a task described as "add vector search to an existing Postgres-backed SaaS app". Substring matching, not full-text: there is no stemming and no ranking, so "migrating" does not match "migration". Max 200 characters and 8 terms; punctuation is treated as a separator and wildcards are not supported.
  • limit · number · optionnel
    Page size (default 20, max 100)
  • cursor · string · optionnel
    Pagination cursor (created_at timestamp from previous page)
  • status · string · optionnel
    Filter by status (comma-separated). Valid values: executing, step_executing, completed, failed, halted, pending_approval, cancelled. E.g. "halted,failed,pending_approval"
  • folder_id · string · optionnel
    Filter by folder ID. Pass "unassigned" for workflows in no folder
  • task_type · string · optionnel
    Filter by classified task type
  • created_after · string · optionnel
    ISO timestamp — only workflows created after this time
  • created_before · string · optionnel
    ISO timestamp — only workflows created before this time
  • counts_toward_limit · string · optionnel · true | false
    Filter to workflows counting toward the concurrent limit

  Exemple d'appel

  {"name": "rigor_workflows", "arguments": {"q": "texte", "limit": 1, "cursor": "texte", "status": "texte", "folder_id": "texte", "task_type": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"ok": true, "workflows": [], "pagination": {}, "credits_remaining": 1, "concurrent_summary": {}}}

C'est votre serveur ?

Revendiquez-le sur l'espace créateur (preuve GitHub, domaine ou npm) pour afficher le badge « propriétaire vérifié » et lancer un pentest comportemental à la demande afin d'obtenir le badge « Vérifié ».

Créer un compte / revendiquer →

Source

https://plith.ai/api/mcp