Tools your AI agent can use.

Which tool to reach for

Start with semantic_locate to find something by meaning, get_context_pack for a bundle around an entity or a question, and find_references for recorded references. Every response carries a freshness and coverage envelope, so read that before acting on an absence: a tool that found nothing and a graph that has not finished answer in the same shape.

The tools

• find_references

  Callers, importers and references of one entity. Chains? trace_data_flow.

  View inputs for find_references

  arguments: answer_only, entity_id, max_chars, query, relation_kinds

• get_context_pack

  Bodies, signatures and routes around entities or a question, in one bundle.

  View inputs for get_context_pack

  arguments: depth, entities, entity_id, focal_body, max_chars, neighbor_bodies, question, token_budget

• get_entity_source

  One entity's exact source, by id.

  View inputs for get_entity_source

  arguments: entity_id | required: entity_id

• graph_neighborhood

  What one entity depends on and what depends on it. Paths: trace_data_flow.

  View inputs for graph_neighborhood

  arguments: depth, direction, entity_id, limit, max_chars | required: entity_id

• impact_analysis

  Every entity a change could affect, from entity_ids or change_ids.

  View inputs for impact_analysis

  arguments: base, change_ids, entity_ids, files, head, max_chars

• kin_graph_status

  Graph and embedding counts. last_settled_selected_graph marks a last settled reading.

• kin_init

  Set a folder up as a Kin repository when it is not one.

  View inputs for kin_init

  arguments: path

• kin_mutate

  Commit targeted entity changes atomically, in one call.

  View inputs for kin_mutate

  arguments: operations, summary | required: operations

• kin_provenance_query

  Who changed an entity, when, and whether it was approved.

  View inputs for kin_provenance_query

  arguments: entity_id, limit | required: entity_id

• kin_session_end

  Session plumbing for writes. Close an open session.

  View inputs for kin_session_end

  arguments: session_id | required: session_id

• kin_session_exec

  Build, test or run this project with its toolchain. Session needs can_execute.

  View inputs for kin_session_exec

  arguments: argv, env, max_output_bytes, session_id, summary, timeout_secs | required: session_id, argv

• kin_session_heartbeat

  Session plumbing for writes. Keep a session alive.

  View inputs for kin_session_heartbeat

  arguments: session_id | required: session_id

• kin_session_start

  Session plumbing for writes. Open one; get a session_id.

  View inputs for kin_session_start

  arguments: capabilities, client_name, cwd, pid, session_id, transport, vendor | required: vendor, client_name, cwd

• kin_transaction_abort

  Transaction plumbing for writes. Discard what is staged and close it.

  View inputs for kin_transaction_abort

  arguments: session_id, transaction_id | required: transaction_id

• kin_transaction_begin

  Transaction plumbing for writes. Open one; get a transaction_id.

  View inputs for kin_transaction_begin

  arguments: scope, session_id | required: session_id, scope

• kin_transaction_commit

  Transaction plumbing for writes. Publish everything staged, atomically.

  View inputs for kin_transaction_commit

  arguments: message, operations, session_id, transaction_id | required: transaction_id

• kin_transaction_stage

  Transaction plumbing for writes. Stage an update, create, delete or rename.

  View inputs for kin_transaction_stage

  arguments: operations, session_id, transaction_id | required: transaction_id, operations

• lexical_lookup

  Exact literals, punctuation included, in stored graph fields. Lexical evidence only.

  View inputs for lexical_lookup

  arguments: cursor, kind, limit, literal, max_chars

• semantic_locate

  Find code by describing what it does. Know the exact name? semantic_search.

  View inputs for semantic_locate

  arguments: cursor, granularity, include_tests, limit, max_chars, query

• semantic_search

  Declarations by name, kind or language. By meaning? semantic_locate.

  View inputs for semantic_search

  arguments: kind, language, limit, max_chars, query | required: query

• trace_data_flow

  The ordered call chain out from one entity. Two endpoints? trace_path.

  View inputs for trace_data_flow

  arguments: compact, depth, direction, focal, include_body, limit_per_step, max_chars, max_response_chars, target | required: focal

• trace_path

  How one entity reaches another, as hops. One endpoint? trace_data_flow.

  View inputs for trace_path

  arguments: direction, from, from_file, limit, max_chars, max_depth, to, to_file | required: from, to