GnosisLab

Overview Schema Source Related Servers Discussions Admin

create_programme

Define a research programme with goal, typed constraints, allowed variables, and budget, then wire the optimizer to steer trials by metric direction.

Instructions

Create a research programme with a goal, constraints, and budget.

Constraints are typed fields, not strings (commitment 4). Wires the optimizer role for this programme.

Structured params (constraints, allowed_variables, budget) may be sent as JSON-encoded strings if your client cannot emit objects.

metric_direction: 'minimize' (e.g. perplexity, loss) or 'maximize' (e.g. accuracy). The optimizer uses this to steer search and rank best trials.

candidate_version_id: optional RSI Phase 0 correlation — which registered candidate created this programme. Must exist if given.

investigation_id: optional zetesis correlation — which open investigation declared this programme as its obligation (open_investigation requires_programme → link_programme). Recorded as claimed provenance: episteme has no read path into zetesis, so the id is stored, not verified.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`goal`	Yes	Research goal — what the programme is trying to learn.
`budget`	Yes	Budget object {max_trials (required), max_wall_time_hours}; extra keys recorded but not enforced. May be a JSON-encoded string.
`constraints`	Yes	Typed constraint fields {gpu_memory_gb, max_training_hours_per_trial, max_parameter_count} — extra keys allowed (commitment 4); may be JSON-encoded.
`investigation_id`	No	Zetesis investigation this programme answers to — claimed provenance, recorded but not verified against zetesis (plan-20261002-1929Z).
`metric_direction`	No	'minimize' (loss-like: perplexity, error) \| 'maximize' (quality: accuracy, F1) — steers search and best-trial ranking.	maximize
`allowed_variables`	Yes	Variables the programme may search/vary; list or JSON-encoded list.
`candidate_version_id`	No	RSI Phase-0 correlation — which registered candidate created this programme; must exist if given.

Output Schema

TableJSON Schema

Name	Required	Description	Default
`status`	No
`programme_id`	No
`budget_extras_not_enforced`	No

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.28Oct 5, 2026

TDQS

A3.5/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the burden and does meaningful work: it discloses that constraints are typed fields rather than strings, that structured params may arrive as JSON-encoded strings, that the call wires the optimizer role, that metric_direction steers search and best-trial ranking, and that investigation_id is stored but never verified against zetesis. It does not state idempotency or what happens on duplicate goal, but the disclosure goes well beyond the schema.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The core action is front-loaded in the first sentence, and each subsequent paragraph covers a distinct concern (constraints typing, optimizer wiring, JSON encoding, metric_direction, two optional correlation ids). Several parenthetical asides ('commitment 4', 'plan-20261002-1929Z') are noise, so it falls short of a clean 5.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists, so return values need not be explained. For a 7-parameter create tool with no annotations, the description covers the two non-obvious optional ids, the metric_direction semantics, and the JSON-string fallback, leaving only minor gaps like side effects or duplicate-name behavior.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so every parameter is already documented in the schema (goal, budget, constraints, investigation_id, metric_direction, allowed_variables, candidate_version_id). The description largely restates those semantics — metric_direction, typed constraints, unverified provenance — adding little not already present. Baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

First sentence gives a specific verb+resource ('Create a research programme') plus the defining fields (goal, constraints, budget), making it easy to separate from list_programmes or close_programme. It does not explicitly name sibling tools, but the create verb is unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description never says when to create a programme versus, say, formulate_hypothesis or design_experiment, nor any prerequisites for use. Only 'Wires the optimizer role for this programme' hints at context, which is a consequence, not when-to-use guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server