RPCS-1 Agent Tuner & Translation Bridge

travisbergen2/rpcs1-sdk

Wraps the RPCS-1 recommendation engine to tune AI agent parameters based on environmental dynamics. Exposes one tool, recommend_agent_configuration, that translates entropy, predictability, stakes, and commitment style into concrete LLM settings like temperature and attention windows. The logic implements the Matching Principle from IMM Paper 9: high-entropy environments get short attention windows, low-entropy get long ones. Useful when you're diagnosing why an agent oscillates between extremes, freezes in ambiguous contexts, or overcommits in volatile situations. The public server at rpcs1.dev/mcp runs read-only with hourly rate limits and OAuth via PKCE. Python and TypeScript SDKs available if you want to embed the logic directly.

tools: 9
transport: HTTP
last updated: Jul 28, 2026

View on GitHub

Install

From registry metadata

$ npx add-mcp https://rpcs1.dev/mcp --name rpcs1-agent-tuner

Run in your terminal. add-mcp asks which agents to add it to (Claude Code, Codex, Cursor and more). Add -g to install it for every project.

Review the command, arguments and environment values before installing. MCP servers run with your local permissions.

Other ways

Claude Code

$ claude mcp add --transport http rpcs1-agent-tuner https://rpcs1.dev/mcp

Run in your terminal. Add --scope user to make it available in every project.

Tools

Verified live · 9 tools

Verified live against the running server on Nov 6, 2026.

recommend_agent_configurationDiagnose why a deployed AI agent may fail. Takes environmental entropy, predictability, stakes, context horizon, and commitment style, then returns receiver profile values (TI, SG, FT, UE, AR), platform parameters (temperature, top_p, strategy), regime prediction, reasoning, a...4 params

Diagnose why a deployed AI agent may fail. Takes environmental entropy, predictability, stakes, context horizon, and commitment style, then returns receiver profile values (TI, SG, FT, UE, AR), platform parameters (temperature, top_p, strategy), regime prediction, reasoning, a...

Parameters* required

environmentobject

target_modelstring

Optional: the actual model id this agent will run on (e.g. "claude-sonnet-4-6", "deepseek-v4-pro"). When it matches a measured per-model receiver entry (E-LIT table), measured translation directive...

target_platformstring

The platform whose runtime parameters should be recommended.one of anthropic · openai · open_source · genericdefault: anthropic

taskobject

interpretDetect ambiguity in user messages using the RPCS-1 Signature Ambiguity Framework. Returns AR level (AR0-AR5), confidence, candidate interpretations with scores, clarifying questions, and suggested next step. Use when a user says something vague, passive-aggressive, or underspe...2 params

Detect ambiguity in user messages using the RPCS-1 Signature Ambiguity Framework. Returns AR level (AR0-AR5), confidence, candidate interpretations with scores, clarifying questions, and suggested next step. Use when a user says something vague, passive-aggressive, or underspe...

Parameters* required

riskstring

Risk category for ambiguity threshold.one of casual · advice · high-stakes · safety-criticaldefault: advice

text*string

The message to interpret.

forkThe calibrated ambiguity surface: deterministic structural fork detectors (reference, scope, grouping, compare-vs-choose, polysemy) with character-offset spans, plus per-reading one-line clarifiers the sender can append to lock a reading in. Returns competing readings, an ask-...2 params

The calibrated ambiguity surface: deterministic structural fork detectors (reference, scope, grouping, compare-vs-choose, polysemy) with character-offset spans, plus per-reading one-line clarifiers the sender can append to lock a reading in. Returns competing readings, an ask-...

Parameters* required

rejectedarray

Reading summaries the user already rejected — never re-offered.

text*string

The message to analyze for forks.

normalizeClean up text with ellipses, fragments, and run-on thoughts into coherent prose. Use when a user types stream-of-consciousness or fragmented input.1 params

Clean up text with ellipses, fragments, and run-on thoughts into coherent prose. Use when a user types stream-of-consciousness or fragmented input.

Parameters* required

text*string

Fragmented text to normalize.

rewriteGet rewrite instructions for adapting text to a specific style: technical, plain, socially_gentle, concise, detailed, or direct. Use when communication needs tone adjustment.2 params

Get rewrite instructions for adapting text to a specific style: technical, plain, socially_gentle, concise, detailed, or direct. Use when communication needs tone adjustment.

Parameters* required

stylestring

Target audience style.one of technical · plain · socially_gentle · concise · detailed · directdefault: plain

text*string

Text to rewrite.

calibrate_profileBuild a ReceiverProfile (TI, SG, FT, UE, AR — continuous 0-100, never a category label) from five behavioral forced-choice answers. Call with NO answers to get the five questions to ask the user; call again with their answers (a/b/c per primitive) to get the profile. Store the...1 params

Build a ReceiverProfile (TI, SG, FT, UE, AR — continuous 0-100, never a category label) from five behavioral forced-choice answers. Call with NO answers to get the five questions to ask the user; call again with their answers (a/b/c per primitive) to get the profile. Store the...

Parameters* required

answersobject

Chosen option id per primitive. Omit entirely to receive the questions.

prepare_promptThe inbound half of the Translation Bridge loop. Takes the user’s raw message (possibly ambiguous, fragmented, or underspecified) plus their ReceiverProfile, and returns the recovered intent, a canonical translation to act on, ambiguity level, and — profile-aware — whether to...3 params

The inbound half of the Translation Bridge loop. Takes the user’s raw message (possibly ambiguous, fragmented, or underspecified) plus their ReceiverProfile, and returns the recovered intent, a canonical translation to act on, ambiguity level, and — profile-aware — whether to...

Parameters* required

profileobject

The user’s ReceiverProfile from calibrate_profile. Shapes clarify-vs-commit behavior.

riskstring

Risk category for the ambiguity threshold.one of casual · advice · high-stakes · safety-criticaldefault: advice

text*string

The user’s raw message.

render_replyThe outbound half of the Translation Bridge loop. Takes your draft reply plus the user’s ReceiverProfile and returns deterministic rendering instructions (structure, warmth, explicitness, revision posture, ambiguity handling — each with a why-trace). Apply the instructions to...2 params

The outbound half of the Translation Bridge loop. Takes your draft reply plus the user’s ReceiverProfile and returns deterministic rendering instructions (structure, warmth, explicitness, revision posture, ambiguity handling — each with a why-trace). Apply the instructions to...

Parameters* required

profile*object

The user’s ReceiverProfile from calibrate_profile.

text*string

Your draft reply.

route_intentEntropy routing over competing interpretations — the model proposes, the deterministic core disposes. YOU generate the candidate readings of the user’s message (3–7 short hypotheses covering the plausible interpretations, INCLUDING likely-typo readings, idiom-vs-literal readin...4 params

Entropy routing over competing interpretations — the model proposes, the deterministic core disposes. YOU generate the candidate readings of the user’s message (3–7 short hypotheses covering the plausible interpretations, INCLUDING likely-typo readings, idiom-vs-literal readin...

Parameters* required

hypothesesarray

Candidate interpretations. Omit to use a generic six-intent starter set plus a catch-all.

likelihoodsobject

Optional externally computed likelihood per hypothesis id (e.g. model-derived) — replaces the lexical scorer.

profileobject

The user’s ReceiverProfile from calibrate_profile. Shapes commit-vs-clarify thresholds.

text*string

The user’s raw message.

Readme

RPCS-1 SDK — AI Agent Tuner

Start with the free tuner: find your AI agent’s likely failure mode, get runtime settings to try, and validate them with a harder case.

RPCS-1 helps teams make agent settings deliberate rather than guessed. Describe the task, change rate, predictability, stakes, relevant context horizon, and commitment style; it returns a five-primitive profile, a runtime recommendation, and a next test. The suite also includes SendRight for catching ambiguous prompts before handoff and the Translation Bridge for profile-aware communication rendering.

Repository Structure

View the full README on GitHub

Details

Repository: github.com/travisbergen2/rpcs1-sdk
Website: rpcs1.dev/tuner?utm_source=official-mcp-registry&utm_medium=mcp-directory&utm_campaign=listing
Category: AI & LLM Tools
Transport: HTTP
Registry: MCP registry, active
Tools verified: Nov 6, 2026

Related AI & LLM Tools MCP servers

View all

Agent Bankwirternow/ai-agent-bank-mcp-serverFinancial infrastructure for AI agents: wallets, USDC transfers, lending, jobs on Polygon0stars

HIPAA Agentxbt786/hipaa-agent-aiHIPAA compliance AI agent — scan, grade, SRA, and generate compliance docs.0stars

Secant Agent Research Packjnilrac/secant-agent-research-mcpPaid MCP web research for agents: search, extraction, citations, and x402 payment discovery.0stars

Memory Nexus Mcpkairosmobius/memory-nexus-mcpPersistent semantic memory for AI agents with knowledge graphs and Intelligence Packs.0stars

Nexus Memory Mcpkairosmobius/nexus-memory-mcpPersistent semantic memory for AI agents with knowledge graphs and Intelligence Packs.0stars

Memory Mcpletta-ai/memory-mcpMCP server for AI memory management using Letta - Standard MCP format0stars