Package: corteza 0.7.1.30

Troy Hernandez

corteza: AI Agent Runtime

An agent runtime that gives Large Language Models (LLMs) from 'Anthropic' <https://www.anthropic.com/>, 'OpenAI' <https://openai.com/>, 'Moonshot' <https://www.moonshot.ai/>, and 'Ollama' <https://ollama.com/> direct access to a live R session with managed workspace state. Tools execute as R function calls with provenance tracking, and a deterministic retrieval system keeps relevant objects in context across turns. Three entry points: a shell command-line interface (CLI), a console read-eval-print-loop via chat(), and a Model Context Protocol (MCP) server via serve() for external clients.

Authors:Troy Hernandez [aut, cre], Grant McDermott [ctb], Jorge Krzyzaniak [ctb], cornball.ai [cph]

corteza_0.7.1.30.tar.gz
corteza_0.7.1.30.zip(r-4.7-any)corteza_0.7.1.30.zip(r-4.6-any)corteza_0.7.1.30.zip(r-4.5-any)
corteza_0.7.1.30.tgz(r-4.6-any)corteza_0.7.1.30.tgz(r-4.5-any)
corteza_0.7.1.30.tar.gz(r-4.7-any)corteza_0.7.1.30.tar.gz(r-4.6-any)
corteza_0.7.1.30.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
corteza/json (API)

# Install 'corteza' in R:
install.packages('corteza', repos = c('https://cornball-ai.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/cornball-ai/corteza/issues

On CRAN:

Conda:

7.47 score 18 stars 6 scripts 235 downloads 86 exports 13 dependencies

Last updated from:7780a08fb6. Checks:7 ERROR, 2 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-develERROR158
source / vignettesOK175
linux-releaseERROR159
macos-releaseERROR121
macos-oldrelERROR130
windows-develERROR145
windows-releaseERROR164
windows-oldrelERROR209
wasm-releaseOK112

Exports:add_observerbot_archive_allbot_configurebot_pollbot_request_flushbot_runbot_run_initbot_run_stepbot_sendchatcontext_limit_for_modelcontext_usage_pctcorteza_execute_in_chatcorteza_execute_in_chat_retaindefault_local_modeldefault_provider_modelensure_skillsestimate_history_tokensestimate_live_context_tokensestimate_text_tokensestimate_tool_tokensformat_ageformat_live_ctxformat_tokensinstall_climatrix_archive_allmatrix_configurematrix_pollmatrix_request_flushmatrix_runmatrix_run_initmatrix_run_stepmatrix_sendmcp_tool_executornew_sessionobserver_progresspolicyregister_skill_from_fnschema_from_fnschema_from_registryservesession_setupskill_installskill_list_installedskill_removeskill_testsubagent_collectsubagent_killsubagent_listsubagent_querysubagent_seed_historysubagent_spawnsubagent_turn_initsubagent_turn_promptsubagent_turn_set_idtool_bashtool_bg_killtool_bg_statustool_cmdtool_collect_subagenttool_exit_plan_modetool_fetch_urltool_git_difftool_git_logtool_git_statustool_grep_filestool_installed_packagestool_kill_subagenttool_list_filestool_list_subagentstool_query_subagenttool_r_helptool_read_filetool_read_handletool_replace_in_filetool_run_rtool_run_r_scripttool_spawn_subagenttool_task_createtool_task_updatetool_web_searchtool_write_fileturnuninstall_clivoice_serveworker_init

Dependencies:callrcodetoolscurldigestjsonlitellm.apiotelprintifyprocessxpsR6sabertinyoauth

Configuration
title: Configuration | Config files | CLI flags | OpenAI-compatible endpoints | JSON config keys | Core | Context | Safety | Skills | Diagnostics | Rate limiting | Subagents | Workspace | Channels | Chat channel | Legacy memory | Slash commands | MCP server | Session tuning | Session parameters | systemd service | Environment variables | R options
Rendered fromconfiguration.mdusingsimplermarkdown::mdweave_to_html

Last update: 2026-08-07
Started: 2026-05-24

Using Codex with corteza
title: Using Codex with corteza | Prerequisites | Authentication | Using Codex on an R package | Provider-plural workflow
Rendered fromcodex.mdusingsimplermarkdown::mdweave_to_html

Last update: 2026-06-10
Started: 2026-06-10

Skills
title: Skills | Format | Frontmatter | Where skills live | How skills get invoked | R skills | Built-in tools | Authoring a skill | Best practices | Sharing
Rendered fromskills.mdusingsimplermarkdown::mdweave_to_html

Last update: 2026-05-27
Started: 2026-05-24

Retroactive-Extraction Runtime
title: Retroactive-Extraction Runtime | Why | Enabling | How it works | What the LLM sees | Permission model | Recursion | Recovery | Summary styles | Known limitations | Disabling
Rendered fromretroactive-extraction.mdusingsimplermarkdown::mdweave_to_html

Last update: 2026-05-26
Started: 2026-05-24

Package as Skill
title: Package as Skill | Setup | 1. Install the package | 2. Register it as a skill | Use | What makes a package a good fit | Why no MCP server | Where to go next
Rendered frompackage-as-skill.mdusingsimplermarkdown::mdweave_to_html

Last update: 2026-05-05
Started: 2026-04-25

Readme and manuals

Help Manual

Help pageTopics
Add a tool-call observer to a sessionadd_observer
Flush all in-memory matrix sessions to the pensar vaultbot_archive_all
Configure the Matrix channel for this hostbot_configure
One iteration of sync-and-replybot_poll
Ask the running matrix bot to archive sessions to pensarbot_request_flush
Run the Matrix adapter as a long-poll loopbot_run
Initialize the Matrix long-poll statebot_run_init
One Matrix long-poll iterationbot_run_step
Send a message to a Matrix roombot_send
Start Interactive Chatchat
Detect the preferred local Ollama modeldefault_local_model
Install corteza CLIinstall_cli
Deprecated: use bot_archive_all()matrix_archive_all
Deprecated: use bot_configure()matrix_configure
Deprecated: use bot_poll()matrix_poll
Deprecated: use bot_request_flush()matrix_request_flush
Deprecated: use bot_run()matrix_run
Deprecated: use bot_run_init()matrix_run_init
Deprecated: use bot_run_step()matrix_run_step
Deprecated: use bot_send()matrix_send
Build a tool executor that routes through an MCP connectionmcp_tool_executor
Create a new turn sessionnew_session
Built-in progress observer that prints to stdoutobserver_progress
Evaluate policy for a tool callpolicy
Start MCP Serverserve
Configure and construct a session for any channelsession_setup
Install a skill from a path or URLskill_install
List installed skillsskill_list_installed
Remove an installed skillskill_remove
Run skill testsskill_test
Collect the result of a previously-fired async subagent query.subagent_collect
Kill a subagent.subagent_kill
List active subagents.subagent_list
Query a subagent.subagent_query
Spawn a subagent.subagent_spawn
Run one agent turnturn
Uninstall corteza CLIuninstall_cli
Serve AgentVoice for live voice clientsvoice_serve