clew-code 0.4.1 → 0.4.3
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.
- package/README.md +23 -10
- package/dist/main.js +1712 -1711
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -138,16 +138,29 @@ A file-backed persistent task queue with lease-based concurrency, exponential-ba
|
|
|
138
138
|
### 70+ built-in tools
|
|
139
139
|
Read, Write, Edit, Glob, Grep, Bash, WebSearch, WebFetch, Browser (Playwright), NotebookEdit, LSP, subagent dispatch (`Agent`), 17 LAN peer-coordination tools, MCP tools, `delegate` (exec/pty to external CLIs), and agent-driven memory curation.
|
|
140
140
|
|
|
141
|
-
### Extensible
|
|
142
|
-
- **MCP** — connect external tools over stdio, SSE (with OAuth), or in-process DirectConnect.
|
|
143
|
-
- **Skills, plugins, hooks** — extend without touching source. Skills via `SKILL.md`, plugins via manifest, and hooks at every lifecycle stage (PreToolUse, PostToolUse, PreBash, PostPrompt, PreAcceptEdit).
|
|
144
|
-
- **8 permission modes** — default, ask, plan, auto, acceptEdits, bypassPermissions, dontAsk, guardian, with granular allow/deny pattern rules.
|
|
145
|
-
|
|
146
|
-
---
|
|
147
|
-
|
|
148
|
-
##
|
|
149
|
-
|
|
150
|
-
Clew Code
|
|
141
|
+
### Extensible
|
|
142
|
+
- **MCP** — connect external tools over stdio, SSE (with OAuth), or in-process DirectConnect.
|
|
143
|
+
- **Skills, plugins, hooks** — extend without touching source. Skills via `SKILL.md`, plugins via manifest, and hooks at every lifecycle stage (PreToolUse, PostToolUse, PreBash, PostPrompt, PreAcceptEdit).
|
|
144
|
+
- **8 permission modes** — default, ask, plan, auto, acceptEdits, bypassPermissions, dontAsk, guardian, with granular allow/deny pattern rules.
|
|
145
|
+
|
|
146
|
+
---
|
|
147
|
+
|
|
148
|
+
## SWE-bench Verified
|
|
149
|
+
|
|
150
|
+
Clew Code ships a SWE-bench Verified evaluation harness modeled after the public
|
|
151
|
+
agent-comparison workflow used by projects like ClawCodex: a FastAPI wrapper
|
|
152
|
+
around `clew -p`, a `prepare`/`run` driver, prediction JSONL output, and optional
|
|
153
|
+
Docker harness execution.
|
|
154
|
+
|
|
155
|
+
No official Clew Code SWE-bench score is claimed in this repository yet. See
|
|
156
|
+
[`eval/README.md`](eval/README.md) to run a smoke batch, full Verified split, or
|
|
157
|
+
side-by-side comparison with another agent.
|
|
158
|
+
|
|
159
|
+
---
|
|
160
|
+
|
|
161
|
+
## Execution layers
|
|
162
|
+
|
|
163
|
+
Clew Code runs work at several layers, each for a different job:
|
|
151
164
|
|
|
152
165
|
| Layer | What it is | Use it for |
|
|
153
166
|
|-------|-----------|------------|
|