devharness 0.8.0 → 0.9.1

package/docs/replay.md

@@ -5,7 +5,7 @@ Record and replay command sequences for testing, automation, and debugging workf
 > **Tip:** Use the `replay-agent` (`.claude/agents/replay-agent.md`) to build sequences through investigation - it records your tool calls automatically.
 >
 > For a condensed, agent-facing version of this material see
-> `skills/cdp-tools/references/sequences.md`. This document is the fuller
+> `skills/devharness/references/sequences.md`. This document is the fuller
 > reference with worked examples.
 
 ## Actions
@@ -139,7 +139,7 @@ Export sequences as Playwright or Puppeteer tests:
 // Export as Playwright test
 replay({ action: 'export', name: 'my-signup-test', format: 'playwright' })
 // Creates: tests/e2e/my-signup-test.spec.ts
-// Also saves: .cdp-tools/sequences/my-signup-test.json
+// Also saves: .devharness/sequences/my-signup-test.json
 
 // Export as Puppeteer test
 replay({ action: 'export', name: 'my-signup-test', format: 'puppeteer' })
@@ -147,7 +147,7 @@ replay({ action: 'export', name: 'my-signup-test', format: 'puppeteer' })
 
 // Export sequence JSON only (default format)
 replay({ action: 'export', name: 'my-signup-test', format: 'sequence' })
-// Creates: .cdp-tools/sequences/my-signup-test.json
+// Creates: .devharness/sequences/my-signup-test.json
 ```
 
 `export` always writes the sequence JSON first, whatever the format. If a target
@@ -159,7 +159,7 @@ and are dropped from the generated test.
 
 ### Configure Export Paths
 
-In `.cdp-tools/config.json`:
+In `.devharness/config.json`:
 
 ```json
 {
@@ -202,7 +202,7 @@ The cursor is only driven for *coordinate* clicks (`input({ action: 'click', x,
 y })`) and for `input({ action: 'press' })`. Selector-based clicks execute
 without a cursor effect.
 
-Configure in `.cdp-tools/config.json`:
+Configure in `.devharness/config.json`:
 
 ```json
 {
@@ -254,7 +254,7 @@ These fields are saved to disk and displayed when listing sequences.
 // Execute commands straight from history without making a sequence
 replay({ action: 'repeat', indices: [12, 13] })
 
-// Execute lines from .cdp-tools/history.log (1-indexed, line 1 = most recent)
+// Execute lines from .devharness/history.log (1-indexed, line 1 = most recent)
 replay({ action: 'runFromLog', lines: [3, 4, 5] })
 ```
 
@@ -304,11 +304,11 @@ replay({ action: 'delete', sequenceId: 'seq-1234567890' })
 ```javascript
 // Export sequence to disk (working directory)
 replay({ action: 'export', sequenceId: 'seq-1234567890', format: 'sequence' })
-// Saves to: .cdp-tools/sequences/<name>.json
+// Saves to: .devharness/sequences/<name>.json
 
 // Export to global location (accessible from any directory)
 replay({ action: 'export', sequenceId: 'seq-1234567890', format: 'sequence', global: true })
-// Saves to: ~/.cdp-tools/sequences/<name>.json
+// Saves to: ~/.devharness/sequences/<name>.json
 
 // List saved sequences on disk (add showAll: true to include completed issues)
 replay({ action: 'listSaved' })
@@ -853,7 +853,7 @@ running too. The run says which it closed:
 #### Declaring the device, not just the browser
 
 `profile` names a persistent Chrome profile (the same ones
-`launchChrome({ profile })` creates, under `~/.cdp-tools/profiles`):
+`launchChrome({ profile })` creates, under `~/.devharness/profiles`):
 
 ```json
 "requiredConnections": [
@@ -1100,7 +1100,7 @@ step counts as a success. A condition that cannot be *evaluated* (bad format,
 unknown type, invalid or over-long regex, tool error) fails the run.
 
 Nesting is capped by `replay.maxConditionalDepth` (default 10) and regexes by
-`replay.maxRegexLength` (default 500); both are `.cdp-tools/config.json`
+`replay.maxRegexLength` (default 500); both are `.devharness/config.json`
 settings. Oscillating chains (A→B→A) are allowed up to the depth limit.
 
 ## forEach Steps
@@ -1316,7 +1316,7 @@ parked on.
 
 ### Configuration
 
-Configure in `.cdp-tools/config.json` (values shown are the defaults):
+Configure in `.devharness/config.json` (values shown are the defaults):
 
 ```json
 {

package/docs/troubleshooting.md

@@ -15,7 +15,7 @@ next call.
 **Solutions:**
 - Relaunch Chrome (`launchChrome`) and restart the dev server
   (`server({ action: 'start', serverId: '...' })`) - its config is kept.
-- Raise or disable the threshold in `.cdp-tools/config.json`:
+- Raise or disable the threshold in `.devharness/config.json`:
   `{"session": {"idleSuspendMinutes": 0}}`. Read at supervisor startup, so it
   applies from the next reconnect.
 
@@ -238,7 +238,7 @@ running until you reboot.
 
 ```javascript
 setDebugLogging({ enabled: true })
-// Check logs at: .cdp-tools/logs/debug.log
+// Check logs at: .devharness/logs/debug.log
 ```
 
 ### Check connection status

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "devharness",
-  "version": "0.8.0",
+  "version": "0.9.1",
   "description": "MCP server that connects AI assistants to Chrome DevTools Protocol for runtime debugging - set breakpoints, inspect variables, monitor network traffic, and automate browser interactions",
   "type": "module",
   "main": "build/index.js",
@@ -56,7 +56,7 @@
   "files": [
     "build/",
     "docs/",
-    "skills/",
+    "plugin/skills/",
     "README.md",
     "LICENSE"
   ],

package/{skills → plugin/skills}/devharness/SKILL.md

@@ -2,7 +2,7 @@
 name: devharness
 description: Drive and debug a running app via the devharness MCP server - launch or attach to Chrome and Node.js, set breakpoints and logpoints, inspect call stacks and variables, watch console and network, manage dev servers, replay any earlier tool call by its history index, and record reproduction sequences that verify a fix. Use whenever a task involves running or debugging a live app, reproducing or verifying a bug, re-driving setup you already did (relaunching, re-logging in, refilling a form), or the user mentions breakpoints, Chrome DevTools, CDP, replay sequences, or devharness tools (launchChrome, navigate, breakpoint, inspect, replay, server, issues, etc.).
 compatibility: Requires the devharness MCP server to be connected (tools such as launchChrome, breakpoint, inspect, replay, server, issues). Previously published as cdp-tools-mcp.
-version: 0.8.0
+version: 0.9.1
 ---
 
 # devharness
@@ -169,8 +169,8 @@ Note this is only one use of `repeat`. The footer hint is on every response, not
 
 If devharness itself seems stuck or broken (not the target app), restart it yourself rather than asking the user to reconnect - don't wait to be told to.
 
-- **Preferred**: `config({ action: 'restart' })`. Under the hood this reads `.cdp-tools/mcp-supervisor.pid` and sends the running mcp-supervisor process a `SIGUSR2`, the same signal `npm run build`'s postbuild hook sends automatically after a rebuild. The supervisor replays the original MCP `initialize` handshake to the freshly spawned child, so the host session never needs to reconnect.
-- If that returns `CONFIG_RESTART_NOT_SUPERVISED` (this server isn't running through the supervisor - e.g. a bare `node build/index.js`), fall back to Bash: `kill -USR2 $(cat .cdp-tools/mcp-supervisor.pid)`.
+- **Preferred**: `config({ action: 'restart' })`. Under the hood this reads `.devharness/mcp-supervisor.pid` and sends the running mcp-supervisor process a `SIGUSR2`, the same signal `npm run build`'s postbuild hook sends automatically after a rebuild. The supervisor replays the original MCP `initialize` handshake to the freshly spawned child, so the host session never needs to reconnect.
+- If that returns `CONFIG_RESTART_NOT_SUPERVISED` (this server isn't running through the supervisor - e.g. a bare `node build/index.js`), fall back to Bash: `kill -USR2 $(cat .devharness/mcp-supervisor.pid)`.
 - If it returns `CONFIG_RESTART_STALE_PID`, the supervisor died without cleaning up its pidfile - ask the user to run `/mcp` to reconnect.
 
 **Expect the triggering call itself to come back as an error - that's normal, not a failure.** In practice `config({ action: 'restart' })` almost never returns its own `CONFIG_RESTART_REQUESTED` success message: the old process gets torn down before it can flush that response, so the supervisor's restart-coordinator answers with a synthesized `MCP error -32000: MCP server is restarting; this request will not receive a response from the previous process. Please retry.` instead. Just retry the next call - it'll hit the freshly restarted (and by then ready) process. Two things to expect on that next call: it runs against a new PID (visible in tool response footers), and any acknowledged monitored-port failures (`server({ action: 'acknowledgePort' })`) reset and may need re-acknowledging, since that state lived in the process that just got replaced.

package/build/session-detector.test.d.ts

@@ -1,9 +0,0 @@
-/**
- * Unit tests for isAgentSession() - the pure classifier used by bug-003's
- * fix (issues({action:'resolve'}) must refuse autonomous agents). It only
- * operates on the plain SessionState shape returned by
- * SessionDetector.getState(), so it's tested directly without any real
- * fs.watch/transcript-file plumbing.
- */
-export {};
-//# sourceMappingURL=session-detector.test.d.ts.map

package/build/session-detector.test.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"session-detector.test.d.ts","sourceRoot":"","sources":["../src/session-detector.test.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG"}

package/build/session-detector.test.js

@@ -1,45 +0,0 @@
-/**
- * Unit tests for isAgentSession() - the pure classifier used by bug-003's
- * fix (issues({action:'resolve'}) must refuse autonomous agents). It only
- * operates on the plain SessionState shape returned by
- * SessionDetector.getState(), so it's tested directly without any real
- * fs.watch/transcript-file plumbing.
- */
-import { describe, it, expect } from 'vitest';
-import { isAgentSession } from './session-detector.js';
-function state(overrides = {}) {
-    return {
-        mainSession: null,
-        activeAgents: [],
-        entryCount: 0,
-        ...overrides,
-    };
-}
-describe('isAgentSession', () => {
-    it('classifies as agent when only an agent transcript has matched this process\'s PID', () => {
-        const s = state({
-            mainSession: null,
-            activeAgents: [{ agentId: 'a1', shortId: 'a1', agentFile: '/agent-a1.jsonl', startedAt: Date.now() }],
-        });
-        expect(isAgentSession(s)).toBe(true);
-    });
-    it('classifies as human when the main session transcript has matched, even with no agents', () => {
-        const s = state({
-            mainSession: { sessionId: 'm1', shortId: 'm1', sessionFile: '/m1.jsonl', detectedAt: Date.now() },
-            activeAgents: [],
-        });
-        expect(isAgentSession(s)).toBe(false);
-    });
-    it('prefers human when both a main session and agents are present (main session wins)', () => {
-        const s = state({
-            mainSession: { sessionId: 'm1', shortId: 'm1', sessionFile: '/m1.jsonl', detectedAt: Date.now() },
-            activeAgents: [{ agentId: 'a1', shortId: 'a1', agentFile: '/agent-a1.jsonl', startedAt: Date.now() }],
-        });
-        expect(isAgentSession(s)).toBe(false);
-    });
-    it('defaults to human (fails open) when nothing has been classified yet', () => {
-        const s = state({ mainSession: null, activeAgents: [] });
-        expect(isAgentSession(s)).toBe(false);
-    });
-});
-//# sourceMappingURL=session-detector.test.js.map

package/build/session-detector.test.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"session-detector.test.js","sourceRoot":"","sources":["../src/session-detector.test.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,OAAO,EAAE,QAAQ,EAAE,EAAE,EAAE,MAAM,EAAE,MAAM,QAAQ,CAAC;AAC9C,OAAO,EAAE,cAAc,EAAqB,MAAM,uBAAuB,CAAC;AAE1E,SAAS,KAAK,CAAC,YAAmC,EAAE;IAClD,OAAO;QACL,WAAW,EAAE,IAAI;QACjB,YAAY,EAAE,EAAE;QAChB,UAAU,EAAE,CAAC;QACb,GAAG,SAAS;KACb,CAAC;AACJ,CAAC;AAED,QAAQ,CAAC,gBAAgB,EAAE,GAAG,EAAE;IAC9B,EAAE,CAAC,mFAAmF,EAAE,GAAG,EAAE;QAC3F,MAAM,CAAC,GAAG,KAAK,CAAC;YACd,WAAW,EAAE,IAAI;YACjB,YAAY,EAAE,CAAC,EAAE,OAAO,EAAE,IAAI,EAAE,OAAO,EAAE,IAAI,EAAE,SAAS,EAAE,iBAAiB,EAAE,SAAS,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC;SACtG,CAAC,CAAC;QACH,MAAM,CAAC,cAAc,CAAC,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IACvC,CAAC,CAAC,CAAC;IAEH,EAAE,CAAC,uFAAuF,EAAE,GAAG,EAAE;QAC/F,MAAM,CAAC,GAAG,KAAK,CAAC;YACd,WAAW,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,OAAO,EAAE,IAAI,EAAE,WAAW,EAAE,WAAW,EAAE,UAAU,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE;YACjG,YAAY,EAAE,EAAE;SACjB,CAAC,CAAC;QACH,MAAM,CAAC,cAAc,CAAC,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;IACxC,CAAC,CAAC,CAAC;IAEH,EAAE,CAAC,mFAAmF,EAAE,GAAG,EAAE;QAC3F,MAAM,CAAC,GAAG,KAAK,CAAC;YACd,WAAW,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,OAAO,EAAE,IAAI,EAAE,WAAW,EAAE,WAAW,EAAE,UAAU,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE;YACjG,YAAY,EAAE,CAAC,EAAE,OAAO,EAAE,IAAI,EAAE,OAAO,EAAE,IAAI,EAAE,SAAS,EAAE,iBAAiB,EAAE,SAAS,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC;SACtG,CAAC,CAAC;QACH,MAAM,CAAC,cAAc,CAAC,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;IACxC,CAAC,CAAC,CAAC;IAEH,EAAE,CAAC,qEAAqE,EAAE,GAAG,EAAE;QAC7E,MAAM,CAAC,GAAG,KAAK,CAAC,EAAE,WAAW,EAAE,IAAI,EAAE,YAAY,EAAE,EAAE,EAAE,CAAC,CAAC;QACzD,MAAM,CAAC,cAAc,CAAC,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;IACxC,CAAC,CAAC,CAAC;AACL,CAAC,CAAC,CAAC"}