Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/bearlike/grove/using-grovenpx skills add bearlike/Grove --skill using-grovegit clone --depth 1 https://github.com/bearlike/GroveWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00198 | $0.04380 |
| Opus 5 | $0.00099 | $0.02190 |
| Sonnet 5 | $0.00040 | $0.00876 |
| Haiku 4.5 | $0.00020 | $0.00438 |
Grade A, and why
using-grove scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 312 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Using Grove
Grove runs a fleet of coding agents that live outside your context rather than sub-agents inside it. Each workspace is a whole separate agent process with its own runtime, its own context window and its own git worktree, addressable from outside. You create, steer, watch, pause, resume, respawn and kill them.
Reach for Grove when you want parallel workspaces that outlive a single turn. For quick in-session parallelism across a few scoped sub-tasks, plain sub-agents are lighter and fit better.
Grove earns its keep on work too big, too parallel or too cross cutting for one agent in one context.
- Decompose and parallelize. Split by component and spin up one workspace per slice. Each gets a clean worktree, so they never touch each other's files.
- Triage at scale. Forty Sentry issues or a whole backlog, one workspace per issue or per cluster. Each agent investigates independently and you review results rather than noise.
- Work one ticket from several angles at once. Implementation, tests and docs run in parallel in separate environments instead of one after another.
- Independent review. Point separate agents, on different models and different harnesses, at the same diff for uncorrelated second opinions.
The model
- One workspace is one isolated runtime for one agent. A dedicated git worktree, or the repo root for a lightweight case, plus a tmux session running that agent's own CLI, on the host or inside a container.
- Model and harness agnostic by design. Every workspace picks its own agent and model. The point is giving each slice the harness that suits it, not stretching one agent across everything.
- You own the fleet and each member owns its own task. Decide the split, spawn one workspace per slice, then poll and steer and collect. A map step and a reduce step, with real isolation so one agent's long tool call never blocks another.
What Grove reports back
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 312 lines · 0 tokens per session scan A 4fa7551c41bd
using-grove is a skill published in the GitHub repository bearlike/Grove (2 stars, last pushed 12d ago), licensed MIT. It adds 198 tokens to every session and 4,380 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
paseo-plugin
Build and manage trusted local Paseo plugins. Use when the user asks to create, edit, install, reload, enable, disable, remove, or troubleshoot a Paseo plugin; add a native surface or sidebar item; use Paseo from plugin code; add plugin RPCs; or contribute composer attachments.
paseo
Paseo reference for managing projects, workspaces, workspace scripts, agents, schedules, and heartbeats.
paseo-help
Answer questions about the Paseo product and app, including setup, configuration, connectivity, providers, workspaces, updates, logs, and troubleshooting. Use when a user inside Paseo asks how Paseo works, how to configure it, or why something is broken; use the paseo skill instead to operate agents and workspaces…
paseo-advisor
Spin up a single agent as an advisor — second opinion on the current task. Use when the user says "advisor", "second opinion", "what does X think", or wants an outside take without delegating the work itself.
paseo-handoff
Hand off the current task to another agent with full context. Use when the user says "handoff", "hand off", "hand this to", or wants to pass work to another agent.
paseo-committee
Form a committee of two high-reasoning agents to step back, do root cause analysis, and produce a plan. Use when stuck, looping, tunnel-visioning, or facing a hard planning problem.