Session Commands Reference

All commands available within an interactive Octomind session. Type the command at the prompt.

Session Management

/help

Show all available commands with descriptions.

/exit / /quit

Exit the current session.

/list [PAGE]

List all saved sessions. Optional page number for pagination.

/session [NAME]

Switch to a different session by name. Without argument, shows current session info.

/clear

Clear the terminal screen.

Information & Monitoring

/info

Display comprehensive session statistics:

  • Token usage (input, output, cached, reasoning)
  • Cost breakdown (per-request and cumulative)
  • Cache savings (tokens and USD)
  • Compression statistics (if compression has occurred)
  • Model information

/report

Generate a detailed usage report for the session with per-request breakdown.

/share

Upload the current session's JSONL log to the share endpoint and print a permanent URL pointing at the web viewer (octomind.run/r/<id>). The full forensic trace — every user/assistant turn, every tool call with args and results, every cost update, every compression/truncation marker — renders in the browser exactly as it occurred on disk.

The CLI does not open the URL automatically — clicking it is your choice.

/share

Output:

url    https://octomind.run/r/<8-char id>
id     <8-char id>

Environment overrides:

  • OCTOMIND_SHARE_URL — point /share and /analyze at a different host (defaults to https://octomind.run). Use this only when pointing at a self-hosted instance or a local dev server.

/analyze

Open the current session in the web viewer without uploading anything. A tiny HTTP server is bound to 127.0.0.1 on a random port; the printed URL points at octomind.run/analyze?b=127.0.0.1:<port>&t=<token> so the browser fetches the JSONL directly from your machine.

The bridge:

  • listens on loopback only — unreachable from other machines,
  • gates every request with a single-use 24-char token sent in the X-Bridge-Token header,
  • aborts the previous bridge when /analyze is re-invoked (fresh port + fresh token each time),
  • shuts down with the octomind process — there is no persistent state and no upload.
/analyze

Output:

url    https://octomind.run/analyze?b=127.0.0.1:<port>&t=<token>
port   127.0.0.1:<port> (loopback only)

Use /analyze for ephemeral, private review of an in-flight session; use /share when you want a permanent link to send to someone else.

/model [MODEL]

Show or change the current model. Without argument, displays current model. With argument, switches to specified model in provider:model format.

/model openai:gpt-4o
/model anthropic:claude-sonnet-4

/role [ROLE]

Show or change the current role. Without argument, displays current role.

/role assistant
/role developer

/effort [LEVEL]

Show or change the reasoning effort level. Without argument, displays current level. With argument, sets the effort to one of: low, medium, high, xhigh, max.

/effort high
/effort max

/loglevel [LEVEL]

Change the log level. Options: none, info, debug.

/loglevel debug

Context Management

/context [FILTER]

View session context (message history). Filters:

  • all -- Show all messages
  • assistant -- Only assistant messages
  • user -- Only user messages
  • tool -- Only tool calls and results
  • large -- Only messages exceeding a token threshold
/context
/context tool
/context large

/done

Mark the current task as complete. Triggers:

  • Plan completion (if a plan is active)
  • Force context compression (preserves only env-loaded skills, drops manually activated ones)
  • Lesson extraction and storage (when [learning] is enabled — see Learning Guide)

Media

/image <PATH>

Attach an image for AI analysis. Supports PNG, JPEG, GIF, WebP.

/image screenshot.png
/image /path/to/diagram.jpg

/video <PATH>

Attach a video for AI analysis.

/video demo.mp4

/copy

Copy the last assistant response to the clipboard.

MCP & Tools

/mcp [ACTION] [ARGS]

Manage MCP servers at runtime.

ActionDescription
/mcp or /mcp listList all MCP servers with status
/mcp infoDetailed server information
/mcp healthForce health check on all servers
/mcp validateValidate MCP configuration
/mcp add <name> <url>Add HTTP server dynamically
/mcp enable <name>Enable a registered server
/mcp disable <name>Disable a server
/mcp remove <name>Remove a server

Commands & Workflows

/run [COMMAND]

Execute a custom command defined in [[commands]] config section. Without argument, lists available commands.

/run reduce
/run estimate

/workflow [NAME]

Execute a workflow. Without argument, lists available workflows.

/workflow developer_workflow

/prompt [NAME]

Inject a prompt template defined in [[prompts]] config section. Without argument, lists available prompts.

/prompt review
/prompt explain

/plan [ACTION]

Manage the structured task plan.

UsageDescription
/plan or /plan showShow current plan with progress

Note: The /plan slash command only displays the current plan. To create, modify, or clear a plan, use the plan MCP tool directly with these commands:

  • plan(command="start", title="...", tasks=[...]) — Create a new plan
  • plan(command="next", content="...") — Mark current task complete, advance to next
  • plan(command="step", content="...") — Add progress note to current task
  • plan(command="reset")Clear/reset the current plan
  • plan(command="done", content="...") — Complete the plan with final summary

/skill [NAME|PAGE|PATTERN]

Manage skills from taps. Skills are reusable instruction packs that inject domain knowledge into context.

UsageDescription
/skillList all skills (active first, then alphabetical)
/skill <name>Toggle a skill on/off
/skill <page>Show page N of the skill list
/skill *pattern*Filter skills by glob pattern

/schedule [SUBCOMMAND] [ARGS]

Direct control over the built-in schedule MCP tool — schedule a message to be injected as a user message at a future time or on the next idle. Same operations as the MCP tool, but driven from chat input. See Scheduled Tasks for the broader use case.

UsageDescription
/schedule or /schedule listList all pending entries with IDs, trigger times, and countdown
/schedule remove <id>Cancel a scheduled entry
/schedule add message="<text>"Schedule a one-shot for the next idle (default when="idle")
/schedule add when="<when>" message="<text>" [every="<interval>"] [description="<label>"]Schedule a new entry
/schedule edit <id> [when="..."] [message="..."] [every="..."] [description="..."]Update an existing entry (use every="none" to clear a repeat)
/schedule helpShow inline usage

Key=value tokens accept shell-style quoting so multi-word values work: when="in 1h 30m", message='hello world'. Supported when formats: idle (fires on next idle — no running taps or background jobs), now (fires immediately), relative (in 5m, in 1h30m, in 90s), time-of-day (15:30, 3:30pm, 9am — tomorrow if past), or absolute (2026-03-30 15:30). every accepts idle (fires on every idle) or the same duration syntax as relative when (10m, 1h, 1h30m). When both when and every are omitted on add, when defaults to idle.

Examples:

/schedule add message="summarize what we just did"             # default: when="idle"
/schedule add when="idle" message="run lint and report results"
/schedule add every="idle" message="remind me to commit"        # fires every idle
/schedule add when="now" message="say the date" every="5m"
/schedule add when="in 5m" message="check the build"
/schedule add when="9am" message="standup reminder" every="1h" description="daily"
/schedule edit abc12345 when="in 1h"
/schedule remove abc12345

/learning [ACTION]

Manage the cross-session learning system. See Learning Guide for full details.

UsageDescription
/learning or /learning statusShow learning system status and statistics
/learning extractForce lesson extraction from current session
/learning injectForce inject relevant lessons into current session