Enum Command 

pub enum Command {
    Scan {
        path: PathBuf,
        verbose: bool,
        quiet: bool,
        exclude_submodules: bool,
    },
    Serve {
        repo: Option<PathBuf>,
        host: Option<String>,
        port: Option<u16>,
        call_log: Option<PathBuf>,
    },
    Status {
        verbose: bool,
    },
    Review {
        no_sync: bool,
    },
    Init {
        client: Option<String>,
        project: bool,
        global: bool,
        dry_run: bool,
        skip_instructions: bool,
    },
    Update {
        check: bool,
    },
    Completions {
        shell: Option<Shell>,
    },
    DebugSnippets {
        path: Option<PathBuf>,
    },
    Decisions {
        command: DecisionsCommand,
    },
    Uninstall {
        client: Option<String>,
        project: bool,
        global: bool,
        dry_run: bool,
    },
}

Top-level subcommands.

Variants

Scan

Scan a project directory and display analysis report.

Fields

path: PathBuf

Path to the project directory to scan.

verbose: bool

Show verbose output: skipped files, detector details, timing.

quiet: bool

Show only errors and final summary.

exclude_submodules: bool

Exclude submodules from scanning (they are scanned by default).

Serve

Start the MCP server for AI agent connections.

Fields

repo: Option<PathBuf>

Repository directory path or project name. Auto-detected from current working directory if omitted.

host: Option<String>

Host to bind the HTTP/SSE transport to (overrides config).

port: Option<u16>

Port for the HTTP/SSE transport (overrides config).

call_log: Option<PathBuf>

Log MCP tool calls to JSONL file for analysis. Default: $XDG_DATA_HOME/seshat/call-log.jsonl

Status

Show indexed projects, submodules, and database info.

Fields

verbose: bool

Show full database paths and additional detail.

Review

Interactive convention review.

On startup, compares the active branch’s last_scanned_commit against git rev-parse HEAD and runs an incremental sync to the current HEAD before opening the TUI, so the review queue reflects the on-disk state.

Fields

no_sync: bool

Skip the pre-TUI freshness check and incremental sync.

Use for emergency / debug access to the existing snapshot when sync would be slow or undesirable. Implies the queue may be stale.

Init

Generate MCP configuration for detected AI clients.

Auto-detects installed AI coding clients. By default uses smart scope: project-level config if it already exists, global config otherwise. For JSON configs, offers to auto-patch with backup. For JSONC, shows a copy-paste snippet.

Fields

client: Option<String>

Specific client to configure. Auto-detects all if omitted. Supported: claude-code, claude-desktop, opencode, cursor

project: bool

Always use project-level configs (in CWD / git root). Writes to .claude/settings.local.json, ./opencode.json, etc.

global: bool

Always use global user configs (default fallback behaviour).

dry_run: bool

Show what would be done without writing any files.

skip_instructions: bool

Only write MCP config; skip agent instructions, skills, and hooks.

Update

Check for newer versions or upgrade the seshat binary.

Fields

check: bool

Only check whether a newer version exists (no installation).

Completions

Print a shell completion script to stdout.

If the shell is omitted, it is auto-detected from the $SHELL environment variable (or PowerShell on Windows). Pipe into the shell’s completion directory or eval it from a shell rc file. Examples:

seshat completions # auto-detect seshat completions bash > /etc/bash_completion.d/seshat seshat completions zsh > “${fpath[1]}/_seshat” seshat completions fish > ~/.config/fish/completions/seshat.fish

Fields

shell: Option<Shell>

Target shell. Auto-detected from $SHELL if omitted.

DebugSnippets

Debug: print all conventions with real evidence snippets from the DB.

Reads conventions from the current project’s database and prints description, nature, confidence, adoption stats, and full snippet text for each evidence item. Use for debugging snippet extraction.

Fields

path: Option<PathBuf>

Path to project directory. Auto-detected from CWD if omitted.

Decisions

Manage user-recorded decisions stored in the project database.

Decisions are project-wide records of approve/reject/partial/recorded outcomes for conventions. They survive branch deletion and are the source of truth for the review queue’s exclusion filter.

Fields

command: DecisionsCommand

The decisions subcommand to execute.

Uninstall

Remove all Seshat configuration from detected AI clients.

Reverses seshat init: removes MCP entries, instruction sections, skill directories, and hook scripts. Does NOT remove the binary or DB files.

Fields

client: Option<String>

Specific client to uninstall. Auto-detects all if omitted. Supported: claude-code, claude-desktop, opencode, cursor

project: bool

Only uninstall from project-level configs.

global: bool

Only uninstall from global user configs.

dry_run: bool

Show what would be removed without making changes.

Trait Implementations

impl Debug for Command

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl FromArgMatches for Command

fn from_arg_matches(__clap_arg_matches: &ArgMatches) -> Result<Self, Error>

Instantiate Self from ArgMatches, parsing the arguments as needed.

fn from_arg_matches_mut( __clap_arg_matches: &mut ArgMatches, ) -> Result<Self, Error>

Instantiate Self from ArgMatches, parsing the arguments as needed.

fn update_from_arg_matches( &mut self, __clap_arg_matches: &ArgMatches, ) -> Result<(), Error>

Assign values from ArgMatches to self.

fn update_from_arg_matches_mut<'b>( &mut self, __clap_arg_matches: &mut ArgMatches, ) -> Result<(), Error>

Assign values from ArgMatches to self.

impl Subcommand for Command

fn augment_subcommands<'b>(__clap_app: Command) -> Command

Append to Command so it can instantiate Self via FromArgMatches::from_arg_matches_mut

fn augment_subcommands_for_update<'b>(__clap_app: Command) -> Command

Append to Command so it can instantiate self via FromArgMatches::update_from_arg_matches_mut

fn has_subcommand(__clap_name: &str) -> bool

Test whether Self can parse a specific subcommand