features added on agent and some fixes on web live tool#7
Closed
GopiB9119 wants to merge 4 commits into
Closed
Conversation
There was a problem hiding this comment.
Pull request overview
This PR is a broad v6.0 update that expands the project from a VS Code-first Playwright Edge MCP toolkit into a more complete “developer automation platform” with stronger safety/verification, additional agent tooling, background execution support, and reorganized documentation.
Changes:
- Adds new JS runtime capabilities and verification/testing (NL parser expansions, structured
runActionoutputs, new smoke-chain test, MCP preflight). - Expands the Python agent tool surface (git/testgen/snapshot/refactor/vision/docs/background task tools), plus redaction hardening and integration-failure artifact capture.
- Reorganizes and adds docs/governance artifacts (new quickstart location, architecture/trust/security guides, CI workflows, schema/action updates).
Reviewed changes
Copilot reviewed 83 out of 86 changed files in this pull request and generated 10 comments.
Show a summary per file
| File | Description |
|---|---|
USAGE_QUICKSTART.md |
Removed root quickstart (moved under docs/). |
tests/smoke-chain.test.js |
Adds end-to-end-style smoke tests for parser → EdgeSession response/verification contract. |
SECURITY.md |
Updates supported-version policy table for v6 line. |
scripts/validate-web-task.js |
Adds a CLI validator for web-task JSON files (schema-adjacent checks). |
scripts/validate-governance.js |
Adds a validator for .github governance structure and frontmatter. |
scripts/resolve-python.js |
Adds helper to locate repo/system Python and run Python modules from npm scripts. |
scripts/mcp-preflight.js |
Adds local MCP/EdgeSession preflight checks and sanity assertions. |
README.md |
Expands product positioning, security model details, commands, and project structure docs pointers. |
playwright-edge-agent.js |
Removes legacy direct-Playwright helper entry point. |
package.json |
Bumps version to 6.0.0; adds daemon-related scripts, validation scripts, and reworks test scripts. |
nl-command-parser.js |
Adds parsing for back/forward/refresh/press/hover/select/focus/clear/double/right click commands. |
index.js |
Introduces an npm entry point exporting EdgeSession/parser/tracing helpers. |
docs/USAGE_QUICKSTART.md |
Adds new consolidated quickstart document under docs/. |
docs/TRUST_RELIABILITY_EXECUTION_PLAN.md |
Adds phased plan for verification, safety, and reliability improvements. |
docs/security_best_practices_report.md |
Adds a security review/report document for recent hardening work. |
docs/RELEASE_CHECKLIST.md |
Updates release checklist with new test/validation commands and CI checks. |
docs/PRODUCT_ARCHITECTURE.md |
Adds unified architecture overview for JS (VS Code) + Python runtimes. |
docs/MEMORY_USAGE.md |
Documents memory tool payloads, shapes, and recommended sequences. |
docs/generate_api_docs.py |
Adds script to generate API_REFERENCE.md from tool schemas. |
Dockerfile |
Adds container build for Node + Python + Playwright Edge runtime. |
cli-agent.js |
Routes CLI execution through runAction and prints sanitized structured results + summary. |
CHANGELOG.md |
Adds changelog including v6.0.0 notes and historical entries. |
agent/agent/web_tools.py |
Ensures web_fetch responses are redacted/sanitized consistently. |
agent/agent/vision_tools.py |
Adds vision tools (encode/compare/describe) with redaction-aware responses. |
agent/agent/tools.py |
Wires many new managers/tools, adds AST-based calculate, hardens call_tool sanitization/recursion rules. |
agent/agent/tooling/schemas.py |
Updates tool schema descriptions and adds include_artifact flags. |
agent/agent/tooling/registry.py |
Registers new tools in the callable allowlist. |
agent/agent/tests/test_workflow_tools.py |
Adds tests for artifact output and nested tool-result sanitization + recursion blocking. |
agent/agent/tests/test_web_tools.py |
Strengthens redaction expectations for fetched content/URLs. |
agent/agent/tests/test_tools.py |
Adds tests for call_tool sanitization/recursion blocking and AST calculator behavior. |
agent/agent/tests/test_oauth_tools.py |
Adds tests ensuring secrets/tokens aren’t exposed in OAuth tool outputs. |
agent/agent/tests/test_mcp_tools.py |
Adds tests for sanitized debug artifact capture (tabs/snapshot/output listing). |
agent/agent/tests/test_fs_tools.py |
Adds tests for secret redaction in FS reads/search/analyze previews. |
agent/agent/tests/test_config.py |
Adds tests for config constants and model-client creation behavior. |
agent/agent/tests/test_command_tools.py |
Adds tests for restricted/permissive command execution and redaction behavior. |
agent/agent/tests/integration/test_mcp_live_integration.py |
Adds failure-artifact capture and improved env/output-dir handling for live tests. |
agent/agent/tests/integration/README.md |
Documents live-test failure artifacts and opt-in raw output copying behavior. |
agent/agent/tests/integration/__init__.py |
Marks integration tests as a package. |
agent/agent/tests/__init__.py |
Marks unit tests as a package. |
agent/agent/SYSTEM_PROMPT.md |
Updates tool list and adds grounding/continuity policy guidance. |
agent/agent/snapshot_tools.py |
Adds snapshot/rollback manager for workspace files. |
agent/agent/refactor_tools.py |
Adds refactor utilities (rename/dead code/duplicates/metrics). |
agent/agent/oauth_tools.py |
Ensures OAuth outputs are redaction-aware and raw token output is blocked by default. |
agent/agent/mcp_tools.py |
Adds sanitized debug-artifact capture; sanitizes args/results/errors consistently; env-configurable trace/session saving. |
agent/agent/doc_tools.py |
Adds documentation tools (docstring templates, changelog entry, doc coverage). |
agent/agent/config.py |
Extracts env/config/model-client setup from agent loop into a dedicated module. |
agent/agent/command_tools.py |
Wraps command tool JSON responses with redaction-aware serialization. |
agent/agent/background/worker.py |
Adds background worker to execute queued tasks using the full toolchain. |
agent/agent/background/task_queue.py |
Adds file-based persistent task queue with locking, progress streaming, retention, recovery. |
agent/agent/background/__init__.py |
Documents background daemon/worker architecture. |
agent/agent/background_tools.py |
Adds foreground tools to submit/monitor/cancel/list background tasks. |
agent/agent/ARCHITECTURE_PLAN.md |
Updates architecture plan status and next steps for v6. |
agent/agent/__init__.py |
Marks agent runtime as a package. |
agent/__init__.py |
Marks top-level agent as a package. |
agent-logic.js |
Removes legacy unsafe browser helper entry point. |
.npmignore |
Adds npm publish ignore rules for dev/CI/artifacts and non-JS-package content. |
.gitignore |
Removes hardcoded user paths; adds ignores for runtime artifacts, venv, snapshots, and eval outputs. |
.github/workflows/python-agent-tests.yml |
Expands CI to run JS check/tests in addition to Python unit tests. |
.github/workflows/nightly-integration.yml |
Adds scheduled nightly job running full suite + MCP preflight + live integration tests with artifact upload. |
.github/workflows/live-web-governance-check.yml |
Adds governance validation workflow for .github + runtime paths. |
.github/skills/web-works/web-task.schema.json |
Expands allowed step actions (fill/hover/search/back/forward/refresh/etc.). |
.github/skills/web-works/SKILL.md |
Reworks skill doc structure and clarifies execution/verification/failure control guidance. |
.github/skills/web-works/PROMPTS.md |
Adds clearer purpose/non-purpose sections and structured brief fields. |
.github/skills/web-works/examples/extract-github-trending.json |
Adds example extract task JSON. |
.github/skills/web-works/examples/automate-wikipedia-search.json |
Adds example automate task JSON. |
.github/README.md |
Rewrites governance guide into “lanes” and clearer ownership/editing guidance. |
.github/prompts/playwright-live-web-task-brief.prompt.md |
Adds structured live-web intake prompt file. |
.github/instructions/playwright-edge.instructions.md |
Expands runtime instruction doc with purpose/non-purpose, evidence, and validation expectations. |
.github/instructions/live-web-governance.instructions.md |
Adds instructions for editing live-web governance assets. |
.github/copilot-instructions.md |
Adds pointers to lane-specific rule sources and clarifies JSON/task handling at high level. |
.github/CODEOWNERS |
Adjusts ownership patterns for workflows/agent/scripts to a single owner. |
.env.example |
Adds a comprehensive example env file with safe defaults and runtime configuration knobs. |
.dockerignore |
Adds docker ignore rules to keep images small and exclude artifacts. |
.devcontainer/devcontainer.json |
Replaces one-line image config with a richer devcontainer setup (Node/Python features, extensions, ports). |
…nded summary check Agent-Logs-Url: https://github.com/GopiB9119/agent-live-web/sessions/1767e2c6-a7a0-4e46-86db-e08827be3dad Co-authored-by: GopiB9119 <93776715+GopiB9119@users.noreply.github.com>
auto-merge was automatically disabled
April 11, 2026 02:26
Head branch was pushed to by a user without write access
…ckerfile, README, and quickstart docs Agent-Logs-Url: https://github.com/GopiB9119/agent-live-web/sessions/335940cc-b616-4b32-9c56-63efd4b2c5bb Co-authored-by: GopiB9119 <93776715+GopiB9119@users.noreply.github.com>
Agent-Logs-Url: https://github.com/GopiB9119/agent-live-web/sessions/335940cc-b616-4b32-9c56-63efd4b2c5bb Co-authored-by: GopiB9119 <93776715+GopiB9119@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.