usebeeline 0.0.120 → 0.0.122

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (3) hide show
  1. package/README.md +15 -5
  2. package/dist/usebeeline.mjs +2071 -955
  3. package/package.json +1 -1
package/README.md CHANGED
@@ -95,7 +95,7 @@ A Room and a corner are the same conversation surface with different permissions
95
95
 
96
96
  **A corner** is a fresh worktree, its own branch, one owning agent, and one fixed objective of at most 24 words. In it the agent works, commits, pushes the branch, runs `gh pr create`, and prints the pull request URL. It then waits for the server's own checks fact — not for whatever `gh` printed locally — and merges only when the checks passed and no human has put the corner on hold. The merge webhook archives the corner and reaps the worktree.
97
97
 
98
- The corner receives a GitHub App token scoped to **that one repository**, installed as a worktree-local git credential helper. Your host credential stores are masked out of the sandbox.
98
+ The corner receives a GitHub App token scoped to **that one repository**, installed as a worktree-local git credential helper. Your host credential stores are masked out of the sandbox — with one deliberate exception: the Trusty Squire session directory stays readable, so a routed agent shares your one signed-in browser and vault instead of standing up its own.
99
99
 
100
100
  **Direct messages** are strictly conversational: no repository binding, no corners.
101
101
 
@@ -104,10 +104,10 @@ The corner receives a GitHub App token scoped to **that one repository**, instal
104
104
  - **The filesystem boundary is the sandbox, not a tool list.** Room sessions run under bubblewrap with a read-only view of the checkout, a private `/tmp`, and an isolated home. Every mounted MCP tool is approved tool-by-tool because the sandbox — not an allowlist — is what holds the line.
105
105
  - **When the sandbox cannot be built, the daemon says so and keeps serving.** A host with no `bwrap`, or a kernel that refuses unprivileged user namespaces, is logged once at start and every session afterwards runs unwrapped; the read-only rule then rests on the harness's own permission callback, which Codex, Claude Code, and Grok honour. Pi does not ask before it writes, so a Pi Room is only as read-only as its sandbox.
106
106
  - **Write access requires a corner.** A corner is a separate worktree on its own branch with a repository-scoped GitHub App token, and it is opened by an explicit host-governed call, never inferred.
107
- - **Reach outside the sandbox is a grant.** The agent asks — `path`, `host`, `secret`, `device`, `budget`, or `command` — and a card goes to its owner in the Room with the exact ask and the reason. You approve once, always, or deny, and the decision is a line in the transcript. Approving a command grant is word-for-word: an approved `npm test` does not approve `npm test && curl …`, and a command carrying shell metacharacters is refused before it is ever offered.
108
- - **Yolo mode** flips a single agent to auto-approval and is settable only by that agent's owner. In a public Workspace, yolo is forced off without changing the owner's preference, so it resumes when the Workspace returns to invite-only; it never covers a budget grant.
107
+ - **Reach outside the sandbox is a grant.** The agent asks — `path`, `host`, `secret`, `device`, `budget`, `command`, or `mcp` — and a card goes to its owner in the Room with the exact ask and the reason. You approve once, always, or deny, and the decision is a line in the transcript. Approving a command grant is word-for-word: an approved `npm test` does not approve `npm test && curl …`, and a command carrying shell metacharacters is refused before it is ever offered. An `mcp` grant routes one MCP server the operator already runs on this host into the agent's isolated home — only that agent's owner can approve it, only for an agent that answers its owner alone (approving on an open agent narrows it to the owner first), and the route mounts on the agent's next session. Revoking it removes the route from the home.
108
+ - **Yolo mode** flips a single agent to auto-approval and is settable only by that agent's owner. In a public Workspace, yolo is forced off without changing the owner's preference, so it resumes when the Workspace returns to invite-only; it never covers a budget or `mcp` grant.
109
109
  - **Provider keys never reach Beeline's servers.** They live in your config directory at mode `0600` and reach only the harness process you already trust with them.
110
- - **Honest about what is not built yet:** today a `command` grant is the one kind that actually changes what a running agent may do. `path`, `host`, `secret`, `device`, and `budget` grants are requested, decided, and recorded, but are not yet applied to the sandbox.
110
+ - **Honest about what is not built yet:** today `command` and `mcp` are the kinds that actually change what a running agent may do. `path`, `host`, `secret`, `device`, and `budget` grants are requested, decided, and recorded, but are not yet applied to the sandbox.
111
111
 
112
112
  ## Command reference
113
113
 
@@ -124,7 +124,8 @@ Runtime state lives in `${XDG_STATE_HOME:-~/.local/state}/beeline/agents/<agent-
124
124
 
125
125
  ## Tool reference
126
126
 
127
- Two MCP surfaces are mounted into every agent session.
127
+ Two Beeline MCP surfaces are mounted into every agent session. Repository-backed Rooms and
128
+ corners also mount the release-owned `codegraph` server after its local index is ready.
128
129
 
129
130
  `beeline-readonly-mcp` — reading, in a Room and in a corner:
130
131
 
@@ -136,6 +137,15 @@ Two MCP surfaces are mounted into every agent session.
136
137
  | `read_agent_file` | Read the agent's approved skills or Workspace memory |
137
138
  | `write_memory` | Replace the agent's private Workspace `MEMORY.md` — the only memory write a Room allows |
138
139
 
140
+ `codegraph` — indexed code relationships in repository-backed Rooms and corners:
141
+
142
+ | Tool | What it does |
143
+ | ------------------- | --------------------------------------------------------------------------------------------- |
144
+ | `codegraph_explore` | Return relevant source, call paths, and blast radius from the repository's generated index |
145
+
146
+ Rooms run CodeGraph without a file watcher and keep source files read-only; only the generated
147
+ `.codegraph` index is writable. Corners keep the watcher so edits are reflected as work proceeds.
148
+
139
149
  `beeline-agent` — acting, host-governed:
140
150
 
141
151
  | Tool | Where | What it does |