CLI Reference¶

Invocation¶

kedi SOURCE [OPTIONS] [PROGRAM_OPTIONS]
kedi -c SOURCE_TEXT [OPTIONS] [PROGRAM_OPTIONS]
kedi --idle [RUNTIME_BACKEND_OPTIONS] [--record] [--load SESSION_PATH] [--no-color]
kedi parse SOURCE
kedi SOURCE --parse
kedi install [PACKAGE.KEDI]
kedi add PACKAGE_NAME
kedi add git+https://github.com/OWNER/REPOSITORY.git
kedi skills add --path SKILL_DIRECTORY
kedi skills add --repo OWNER/REPOSITORY
kedi a2a serve SOURCE --entry PROFILE [OPTIONS]
kedi notebook [--host HOST] [--port PORT] [--python PATH] [--cwd PATH] [--token TOKEN] [--no-open]

Use kedi --help, kedi install --help, or kedi add --help for the relevant usage surface.

-c/--command supports direct execution and --parse only. It cannot be combined with tests, evals, or optimization because those modes require a source-file identity and adjacent artifacts.

Runtime Backend Options¶

Option	Default	Meaning
`--adapter NAME`	`pydantic`	`pydantic`, `dspy`, `langchain`, `claude`, `codex`, `acp`, or `a2a`
`--adapter-model MODEL`	`groq:qwen/qwen3-32b`	Default model identifier; examples explicitly select `openai:gpt-5.6-luna`

The historical option name is --adapter, but it accepts both framework adapters and harnesses. In Kedi source and the Python API, adapter and agent remain separate type-safe concepts.

Interactive Mode¶

kedi --idle starts a process-local incremental session. It uses +++ for a new fragment and ... for continuation input. Enter an empty continuation line to execute a buffered multiline fragment. :multiline opens a one-shot editor that submits on a complete double Enter; Alt+Enter forces submission. :show <expression> inspects a value, help displays command help, and :exit, Ctrl+C, or Ctrl+D closes the session.

:dump writes a strict snapshot and prints kedi --idle --load <session_path>. --record performs that dump automatically before the REPL exits. --load restores the supplied snapshot and keeps recording subsequent changes to the same file.

Live Kedi and embedded-Python syntax highlighting is enabled by default while preserving the shared REPL history. --no-color disables input highlighting and REPL error colors; a nonempty NO_COLOR environment variable does the same. Redirected input or output uses plain line input. The existing --highlight flag remains supported. Interactive mode accepts --adapter and --adapter-model. It rejects a source file, -c/--command, program arguments, --parse, --test, --eval, and --optimize. --record, --load, --highlight, and --no-color require --idle. See Interactive Execution for state persistence, native results, imports, history, and complete terminal behavior.

Validation and Generation Options¶

Option	Meaning
`--test`	Run all `@test:` cases and stop
`--eval`	Run all `@eval:` suites and stop
`--optimize`	Optimize every eligible `> optimize:` span before continuing
`--no-cache`	Disable/remove `> auto:` codegen cache for this command
`--quiet`	Suppress codegen and GEPA progress
`-p`, `--parse`	Parse source without compiling or executing

If both --test and --eval are given, tests run and return first. Avoid the ambiguous combination and invoke each validation mode separately.

Code Generation Options¶

Option	Default
`--codegen-agent`	`pydantic_ai`
`--codegen-model`	`openrouter/minimax/minimax-m2.7`
`--codegen-retries`	`5`

The alternate codegen agent is mock, intended for deterministic tests. Retries count generation stages. Generated implementations are cached beside the source as SOURCE.cache.kedi.

Optimizer Options¶

Option	Default	Meaning
`--optimizer`	`mock`	`mock` or `gepa`
`--optimizer-model`	`openrouter/minimax/minimax-m2.7`	GEPA main model
`--optimizer-reflection-model`	same	GEPA reflection model
`--optimizer-max-metric-calls`	`100`	Optimization evaluation budget
`--optimizer-model-max-tokens`	`8000`	Main-model output limit
`--optimizer-reflection-model-max-tokens`	`32000`	Reflection output limit
`--optimizer-reflection-minibatch-size`	`3`	Examples before reflection
`--optimizer-max-validation-examples`	all	Explicit validation-set cap
`--optimizer-fresh`	false	Delete prior optimized state/checkpoints

--optimize does not imply GEPA; the CLI default optimizer is mock. --optimizer-fresh removes .optimized.json, .optimized_scores.json, and .gepa/ artifacts. It does not control codegen cache.

Program Arguments¶

Unrecognized options after the source are parsed into the reserved args binding:

kedi report.kedi --incident-id 42 --dry-run

[incident_id: int] = `int(args.incident_id)`
[dry_run: bool] = `args.dry_run is True`

Hyphens normalize to underscores. --key value stores a string; a flag without a value stores True. The first duplicate option wins. Non-option extra tokens are ignored. Missing attributes read as None.

Package Commands¶

kedi install accepts zero or one manifest path and defaults to ./package.kedi. kedi add accepts exactly one registry name or credential-free GitHub git+https URL. Package subcommands reject ordinary runtime options.

A2A Server¶

kedi a2a serve SOURCE --entry PROFILE exposes one exported profile through the official A2A protocol server. Install kedi[a2a] first.

Option	Default / meaning
`--entry NAME`	Required exported profile
`--host`, `--port`	`127.0.0.1`, `8000`
`--public-url URL`	Required for a non-loopback bind
`--rpc-path PATH`	`/a2a`
`--auth`	`basic`, `bearer`, or `api-key`; default `basic`
`--username`	Basic username; default `kedi`
`--secret-env`, `--secret-file`	Secret reference; mutually exclusive
`--api-key-header`	API-key header; default `X-API-Key`
`--allow-unauthenticated`	Explicit loopback-only development mode
`--model`	Server entry model override
`--max-concurrency`	Concurrent profile executions; default `4`
`--max-pending-tasks`	Global and per-principal pending bound
`--max-sessions`	In-memory principal/context session bound
`--max-request-bytes`	HTTP request body bound
`--max-input-chars`	Task text bound
`--max-response-bytes`	Result bound
`--max-tasks-per-principal`	Retained task bound

See A2A Cloud Agents for authentication, structured results, lifecycle recovery, and deployment limits.

Notebook Command¶

kedi notebook dispatches to the separately installed kedi-notebook package. The source checkout supports uv run --extra notebook kedi notebook once its notebook submodule is initialized. See Local Notebook.

Option	Default / meaning
`--host`	`HOST` or `127.0.0.1`
`--port`	`PORT` or `8788`
`--python PATH`	Additional base interpreter; repeatable
`--cwd PATH`	Current directory; import and host command base
`--token TOKEN`	`KEDI_NOTEBOOK_TOKEN`; required for non-loopback serving
`--no-open`	Do not open the browser automatically

Use kedi-notebook --help for the notebook server's full option list. These are notebook options, not flags for ordinary kedi program.kedi execution.

Skill Installation¶

kedi skills add requires exactly one of --path or --repo. The source must provide a root SKILL.md. Installation copies that skill into $KEDI_HOME/registry/skills; it does not activate it in a program. See Skills for loading and trust boundaries.

Exit and Output¶

Success exits 0. Parse, compile, execution, validation, codegen, optimization, package, and usage failures exit nonzero and render source-aware diagnostics when available. Test mode exits 1 when any case fails; eval mode exits nonzero for suite/configuration failures.

Ordinary top-level return values are written to stdout. Diagnostics and failed operation details use stderr. --quiet does not hide final results or errors.