tool.json

{
 "name": "srdcheck",
 "version": "0.9.0",
 "description": "Deterministic rules verdicts over the SRD for game-running agents. Exit codes are scoped results: 0 passes the named checked scope, 1 conflicts with it, 2 cannot-adjudicate (unknown content or a DM-authority decision, including when the caller is the agent-DM — an honest refusal, not an error). Applied rules carry citations; boundary refusals do not invent them.",
 "status": "alpha; versioned machine schemas and the N/N-1 semantic window are reported by `capabilities`",
 "rulesets": [
  {
   "name": "srd-5.2.1",
   "version": "0.2.1",
   "data_version": "0.2.1",
   "rules_version": "0.2.1"
  },
  {
   "name": "toy-tictactoe",
   "version": "0.1.0",
   "data_version": "0.1.0",
   "rules_version": "0.1.0"
  }
 ],
 "invocation": {
  "command": "python -m srdcheck",
  "subcommands": {
   "jurisdiction <name>": "Is this named entity known to the loaded rulesets? exit 0 = known content (all matching categories in data.categories); exit 2 = unknown/third-party content.",
   "cite <heading>": "Verbatim, page-numbered source text for a named SRD heading. Provenance surface: returns the page's own text, never interpretation.",
   "query <type> <params-json>": "Adjudicate a structured proposal. Query types are supplied by the loaded adapters; see query_types below. Existing query tools accept optional table_policy and policy_context metadata.",
   "query <type> <params-json> --table-evaluation": "Project the scoped verdict into deterministic, self-attested table.evaluation/1.0. Add --table-context JSON for caller-owned session/entity/correlation references. Table and execution authority stay external.",
   "--pipe": "Read one {\"type\", \"params\"} JSON query from stdin.",
   "--schema": "Print JSON Schema for input and output.",
   "capabilities": "Print engine, protocol, adapter versions/digests, query types, refusal recovery and observability contracts, and MCP tool names.",
   "query ... --trace [--request-id <id>]": "Keep the verdict on stdout and emit metadata-only lifecycle NDJSON on stderr.",
   "policy validate|export <manifest>": "Import and validate a portable srdcheck.table-policy/1.0 JSON manifest, or export its canonical human/machine form."
  },
  "query_types": [
   "attack.modifiers",
   "check.make",
   "concentration.check",
   "creature.stats",
   "creature.valid",
   "encounter.xp-budget",
   "event.apply",
   "feature.uses",
   "grapple.initiate",
   "help.assist",
   "mage-hand.use",
   "opportunity-attack.provoked",
   "passive.perception",
   "reaction.available",
   "roll.compose",
   "save.check",
   "spell.facts",
   "transition.commit",
   "ttt.move",
   "ttt.options",
   "turn.options",
   "turn.plan"
  ]
 },
 "output": "JSON verdict object on stdout: {verdict, exit_code, why, citations[], rule_ids[], adapter, data?}. Exit-2 data contains stable reason and recovery fields. Schema and refusal contract identities are published by --schema, MCP outputSchema, and capabilities.",
 "table_evaluation": "query/--pipe --table-evaluation maps legal to checked_clean (or checked_with_advisories when a table policy applies), illegal to an exact-evidence finding, and structured refusals to invalid, unsupported, or incomplete. Output is always self_attested. Its named-query scope is machine-readable and every finding evidence ref resolves in its effective policy. Stateful queries expose cursor.state_precondition_hash.",
 "notes": "The 'why' field is templated explanatory prose, not a machine compatibility field. Verdicts are deterministic: same query, same answer, every time. On cannot-adjudicate, branch on structured recovery data rather than prose; an authorized agent-DM may resolve a DM-authority table ruling directly and reuse it through a caller-owned table-policy manifest. turn.options and turn.plan are consistency-tested against each other. For state writes, evaluate with event.apply and verify the exact proposal with transition.commit; the host owns atomic persistence. Use `capabilities` for checked/unchecked scope, refusal mappings, the metadata-only observability schema, and the exact release tuple.",
 "mcp": {
  "command": "python3 -m srdcheck.mcp",
  "transport": "stdio",
  "protocolVersion": "2025-06-18",
  "tools": [
   "jurisdiction",
   "table_evaluation",
   "attack_modifiers",
   "check_make",
   "concentration_check",
   "creature_stats",
   "creature_valid",
   "encounter_xp_budget",
   "event_apply",
   "feature_uses",
   "grapple_initiate",
   "help_assist",
   "mage_hand_use",
   "opportunity_attack_provoked",
   "passive_perception",
   "reaction_available",
   "roll_compose",
   "save_check",
   "spell_facts",
   "transition_commit",
   "turn_options",
   "turn_plan",
   "ttt_move",
   "ttt_options"
  ],
  "notes": "Zero-dependency stdlib MCP server. Tool descriptions/schemas are supplied by loaded adapters (queries.json). Verdicts arrive as structuredContent; illegal and cannot-adjudicate are results, not protocol errors. The opt-in table_evaluation tool projects one native query without changing native tool response schemas. Set SRDCHECK_TRACE=stderr and pass an optional request_id for out-of-band metadata-only tracing."
 },
 "_generated_by": "scripts/gen_tool_json.py"
}