14 tools from the Plith MCP Server, categorised by risk level.
View the Plith policy →burnrate_budget Get today's tracked LLM spend, per-model breakdown, projection, and budget alerts. Free — no credits charged. burnrate_estimate Before executing a multi-step agent plan, estimate the total LLM cost. Returns per-step breakdown and optimization suggestions. burnrate_track Call after LLM invocations to record actual token usage. Cost is computed server-side using current provider pricing — no manual rate lookup needed... dedupq_check Before executing any LLM task, check if an identical or semantically similar task has already been completed. Returns cached result on hit, saving ... guardrail_check Evaluate a proposed agent action against governance policies. Returns allow or deny. Deterministic rule evaluation — no LLM. guardrail_list_policies List all active governance policies for your org, ordered by priority (lowest number first). Use to audit current rules before creating new policie... pitfalldb_query Check for known failure patterns before executing a task type. Returns pitfalls with severity, fix suggestions, and confidence scores. pitfalldb_report Report an agent failure. PII-scrubbed before storage. Linked to existing pitfalls if similar. Free — no credits charged. pitfalldb_stats Get aggregate statistics: total pitfalls, total reports, top task types, recent activity. Free — no credits charged. qualitygate_validate Validate agent output against your rules. Deterministic checks (regex, JSON schema, syntax) plus optional LLM-powered tone and factual analysis. burnrate_optimize Get a cheaper equivalent plan by substituting models with lower-cost alternatives. Call after burnrate_estimate if the estimated cost exceeds your ... dedupq_complete After executing a task, store the result so future identical or similar tasks return a cache hit. guardrail_create_policy Create a reusable governance policy that guardrail_check evaluates deterministically (no LLM) on every call. Use this to define rules before callin... The Plith MCP server exposes 14 tools across 3 categories: Read, Write, Execute.
Use Intercept, the open-source MCP proxy. Write YAML rules for each tool — rate limits, argument validation, or deny rules — then run Intercept in front of the Plith server.
Plith tools are categorised as Read (10), Write (3), Execute (1). Each category has a recommended default policy.
Deterministic policy on every MCP tool call. Per-identity grants. Full audit log.