CLI Reference

Opens the Interactive Control Center — a full-screen TUI for managing projects, configuring providers, and launching scans without memorizing flags.

Runs the full 5-phase agent loop on the current directory. If a prior run exists, only files changed since the last Git commit are re-processed.

Interactive wizard that securely stores your LLM provider, model selection, and API keys. Runs once — no need to export env vars on every session.

List, add, rotate, or delete stored API keys for any configured provider.

Removes CODEBASE.md, links.json, .codilay_state.json, and any cached conversation data for the given project path.

Reports how many files are documented, how many sections exist, and which sections are marked stale (source changed since last doc run).

Git-level diff showing which source files were modified, added, or deleted since the last codilay run.

Opens an interactive chat session. Three escalation layers: doc-based (instant), deep source reading (when doc can't answer), and a learning loop that patches the doc with new answers.

TF-IDF inverted index search across every past chat session for the project. Finds that retry logic discussion from two weeks ago in milliseconds.

Starts the FastAPI server and opens the 3-layer web documentation browser: Reader (static, no LLM), Chatbot (doc context), and Deep Agent (reads source when needed).

Runs CodiLay as a background daemon. Listens for filesystem events via watchdog and triggers incremental re-runs on save. Debounced to avoid redundant work.

Produces a compressed, token-efficient export of your documentation tailored for LLM context windows. Supports LLM-guided interactive customization, natural language queries, and pre-configured presets.

Compares the current documentation against the snapshot from the previous run, showing exactly which sections were added, removed, or modified. Snapshots are saved automatically after every run.

Generates a focused "what changed" report instead of a full reference. Feeds git diffs and existing context to the AI to explain the impact of changes since a specific commit, tag, date, or branch.

Performs targeted security, architecture, performance, or quality scans. Analyzes the wire graph and file content through 60+ specialized audit lenses.

Renders and filters the full wire dependency graph. Slice by wire type, file layer, module, connection count, or direction to surface architectural hubs and reduce noise on large repos.

Record a correction for a file or glob pattern that was mis-classified during triage. Corrections are stored per-project and automatically injected into future triage prompts.

Shows all recorded triage corrections for the project.

Attaches a free-text hint to a project type (e.g. react, django) that is injected into the triage prompt to guide classification.

Deletes a stored triage correction for a specific file path.

Wipes all stored triage corrections for the project. Requires --yes confirmation flag.

Registers a team member in the project's shared knowledge base. Used to attribute facts, decisions, and annotations.

Stores a shared fact about the project. Facts are injected into LLM context during documentation and chat sessions.

Logs an architectural decision record (ADR) with rationale. Surfaced to the AI during documentation runs.

Records a coding convention that the AI will respect when generating documentation and answering chat questions.

Attaches a team note to a specific file, optionally scoped to a line range. Visible to the AI when processing that file.

List stored facts, decisions, conventions, file annotations, or team members. All support filtering flags.

Upvote or downvote a stored fact by its ID. Vote scores are surfaced to the AI to indicate community confidence.

Sets up automatic documentation re-runs triggered by a cron expression, new commits on a branch, or both. Runs as a background daemon with PID file management.

Starts the configured scheduler in the foreground.

Stops the running scheduler daemon for the project.

Displays the current schedule configuration and daemon status.

Disables the schedule without deleting it. Re-enable with schedule set.

Uses wire knowledge to write docstrings and wire comments directly into your source code. Language-aware and supports multiple annotation levels.

Reads the git diff for commits and writes a plain-language document explaining what changed in each file and why it matters. Optionally runs a second pass to score the diff across five quality dimensions.

Install or remove git hooks, such as automatically generating commit docs after each commit.