Menu
Search, browse, and read agent conversation transcripts across Claude, Codex, Gemini, and OpenCode. Use this skill to find previous sessions, recover context, or inspect what agents have done.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Gotchas for the agents-cli headless release. The macOS home base signs, notarizes, and npm-publishes over SSH with no GUI and no Touch ID — three things must be primed or the release stalls. Triggers on: release stuck at sign/publish, errSecInternalComponent, "no npmjs.com bundle", "Missing embedded.provisionprofile", making the release autonomous.
Organize AI coding agents into teams that collaborate on a shared task. Create teams, add teammates, start them, monitor progress, and collect results. Use this skill when you need parallel agent execution. For single-agent dispatch, use `agents run` instead.
Use FIRST for any question about THIS codebase's structure, symbols, or relationships — where a function/class/file is defined, what calls or imports it, what breaks if you change it, how two parts connect, or to load focused source context before editing or reviewing. Routes to the cgraph MCP tools (graph_query / graph_explain / graph_impact / graph_path / graph_context), which serve ranked file:line results, node neighborhoods, transitive blast radius, and token-budgeted source bundles from a resident per-project graph daemon in ~10ms — far cheaper than grepping and reading files. Prefer over blind grep/read for code navigation, dependency tracing, and impact analysis.
Register and connect to your machines over Tailscale SSH with agents-cli. Use this skill to sync devices from the tailnet, list them, open a shell on another machine, or see agent sessions running across your whole fleet.
Schedule agents to run on a cron schedule or one-shot at a specific time. The scheduler auto-starts on first add. Triggers on: 'schedule an agent', 'recurring job', 'cron', 'daily', 'every weekday', 'run at 2pm'.
Execute a single agent headlessly or interactively. Supports plan/edit/auto/skip modes, secrets bundle injection, version pinning, fallback chains, balanced rotation, profile dispatch (Kimi/DeepSeek/etc.), and workflow dispatch by name. Triggers on: 'run claude', 'run codex', 'agents run', 'dispatch an agent', 'headless agent', 'one-off agent task'.
| name | sessions |
| description | Search, browse, and read agent conversation transcripts across Claude, Codex, Gemini, and OpenCode. Use this skill to find previous sessions, recover context, or inspect what agents have done. |
| argument-hint | [search query or session ID] |
| allowed-tools | Bash(agents sessions*), Bash(agents cost*), Bash(agents logs*) |
| user-invocable | true |
Search and browse agent conversation transcripts. This skill teaches you how to use the agents sessions CLI effectively.
# Interactive picker: browse and search recent sessions
agents sessions
# List sessions from current project
agents sessions | head -20
# Search sessions by text
agents sessions "add auth middleware"
# Filter by project across all directories
agents sessions --project agents-cli --all
| Filter | Example | Description |
|---|---|---|
--agent | --agent claude | Filter by agent type |
--all | --all | Include sessions from every directory |
--project | --project myapp | Filter by project name |
--since | --since 2h | Only sessions newer than this |
--until | --until 2026-01-01 | Only sessions older than this |
--limit | --limit 10 | Maximum sessions to return |
--sort | --sort cost | Order by cost, duration, or recent (default) |
--active | --active | Only currently running sessions |
--teams | --teams | Include team-spawned sessions |
# Render session as markdown
agents sessions --markdown <session-id>
# Output as JSON
agents sessions --json <session-id>
# Include only specific roles
agents sessions --markdown --include user,assistant <session-id>
# Show only first/last N turns
agents sessions --markdown --last 10 <session-id>
# List all files written or edited during a session
agents sessions --artifacts <session-id>
# Read a specific artifact
agents sessions --artifact <filename> <session-id>
# Live-tail a session file (Claude and Codex only)
agents sessions tail <session-id>
# Press Ctrl+C to stop
# Or the unified viewer: resolves a session id OR a host-dispatch run (from
# `agents run --host`), and -f follows either (a session tail here is
# `sessions tail` under the hood)
agents logs <id> # show the transcript / run log
agents logs <id> -f # follow a live one
Every session is priced offline at scan time (costUsd) and its wall-clock
runtime is stored (durationMs). Both appear in --json output, and the list
can be sorted by either.
# The 10 most expensive sessions, anywhere
agents sessions --all --sort cost --limit 10
# Longest-running sessions in this project
agents sessions --sort duration
# Per-session cost in JSON
agents sessions --all --sort cost --json | jq '.[] | {shortId, agent, costUsd, durationMs}'
Use agents cost for a fleet-wide rollup (daily $ histogram, top sessions,
per-agent/project/day breakdown):
agents cost # daily histogram + top-10 + per-agent breakdown
agents cost --since 30d # last 30 days
agents cost --by project # group by project instead of agent
agents cost --by day --json # machine-readable daily rollup
agents cost answers "how much did this cost?"; agents usage answers "how
much quota / rate limit is left?" — different commands.
--active to find sessions running right now across terminals, teams, cloud, and headless agents--teams to see what team-spawned agents are doing--since 1h for recent activity--sort cost / --sort duration to find your priciest or longest sessionsagents cost for spend rollups; agents usage for rate-limit statusagents sessions --project myapp --since 1d --agent claude