Releases: kalil0321/reverse-api-engineer
v0.9.0
Added
- Cursor SDK support via
sdk=cursor/--sdk cursor, backed by a bundled Node bridge around the Cursor TypeScript SDK. - Cursor bridge package files are shipped without
node_modules; dependencies install on first Cursor use.
Fixed
- Manual REPL model resolution for Cursor, OpenCode, and Copilot SDK selections.
- Cursor streaming buffers model deltas into coherent blocks and avoids stray
..output / bridge hangs. - Restored the sync temporary-file helper used by the existing sync test surface.
Removed / Security
- Removed the experimental Chrome extension/native host workspace and related CLI subcommands.
- Dropped the
[pricing]LiteLLM extra and included dependency security bumps listed in the changelog.
v0.8.0 - Agent-friendly CLI
Highlights
This release makes reverse-api-engineer invocable from other agents and scripts via a structured non-interactive surface, and retires two providers and the legacy tag system.
Added
- Agent-friendly CLI:
--json,--no-interactive,--dry-run,--headless,--json-schema-version v2 - Stable JSON contract: normalized usage shapes and
error_kindenum for machine-readable error handling engineer --prompt/--fresh: replace@id <run_id> [--fresh] <prompt>REPL syntax with first-class flags- Expanded
--helpepilogs with examples on every subcommand
Changed
- Prompts extracted to markdown templates under
src/reverse_api/prompts/for easier review - New painted-ruins JPG banner
Removed (Breaking)
browser-useandstagehandagent providers: upstream churn made them unreliable. Useauto(Playwright MCP) orchrome-mcp(Chrome DevTools MCP). The[agent]extra is gone; configs auto-migrate toauto- Tag system (
@record-only,@codegen,@id,@docs,@help): replaced by CLI flags.@codegenand@docsremoved entirely (the latter returns later as adocssubcommand) agent --reverse-engineer/--no-engineer: was parsed but never wired through. Usemanual --no-engineerfor HAR-only recordings
Fixed
engineer --jsonemits JSON even whenRUN_IDis missing- Follow-up prompt suppressed in
--json/--no-interactivemode
📦 PyPI: https://pypi.org/project/reverse-api-engineer/0.8.0/
📜 Full changelog: https://github.com/kalil0321/reverse-api-engineer/blob/v0.8.0/CHANGELOG.md
v0.7.1
v0.7.0
What's New
run command
Execute generated scripts directly from the CLI:
reverse-api-engineer run ashby # fuzzy match by name
reverse-api-engineer run a450e520ca30 # by run ID
reverse-api-engineer run ashby --ls # list scripts
reverse-api-engineer run ashby --file api_client.py
reverse-api-engineer run ashby -- --org acme # pass args to script- Shared venv at
~/.reverse-api/runs/.venvwithrequestspre-installed - Auto-install missing imports — offers to
pip installand retry onModuleNotFoundError - Real-time output — stdout streams live, stderr captured for error detection
Security
- litellm >=1.83.0 (critical + high)
- requests >=2.33.0 (medium)
- aiohttp >=3.13.4 (4 medium + 6 low)
- pygments >=2.20.0 (low)
- cryptography >=46.0.6 (low)
Install / Upgrade
pip install -U reverse-api-engineer
# or
uv tool upgrade reverse-api-engineerv0.6.0
See CHANGELOG.md.
PyPI: reverse-api-engineer 0.6.0
Highlights
- Chrome DevTools MCP (
chrome-mcp) as an agent provider next to Playwright MCP (auto) - Smaller sdist on PyPI (excludes local demo video, packed extension zip, store screenshots)
- Streaming fix: tool results fall back to
result/outputwhencontentis empty
Install: uv tool install reverse-api-engineer==0.6.0 or pip install reverse-api-engineer==0.6.0
v0.5.0 - Follow-up chat, abort, AskUserQuestion free mode
What's New
Follow-up Chat
After a run completes, type follow-up messages to iterate in the same session — no new run IDs or folders. The agent retains full conversation context. Press Enter to finish and return to the REPL.
Abort Run (Ctrl+C)
Gracefully cancel a running agent/engineer session with Ctrl+C. Returns to the REPL instead of exiting the app.
AskUserQuestion Free Mode
All select/checkbox prompts now include "Other (type your answer)" so you can always provide free-text input. The agent prompt now documents free-form, multi-select, and multi-question capabilities.
Random Task Suggestions (Ctrl+R)
Press Ctrl+R in agent mode to fill the prompt with a random curated task idea. Press again to cycle through 20 examples.
Fixes
- Usage tracking: Token counts now accumulate across all follow-up turns instead of being overwritten
- Agent mode follow-up: Auto engineer now reuses the shared conversation loop
Install / Upgrade
uv tool install reverse-api-engineer --upgrade
# or
pip install reverse-api-engineer==0.5.0Full Changelog: v0.4.5...v0.5.0
v0.4.5 - Fix stream closed errors
Fixed
- Stream closed errors (#51): Reverted from
query()toClaudeSDKClientwhich maintains a persistent bidirectional connection, eliminating "Error in hook callback hook_0: Stream closed" errors - CLAUDECODE env var leak: Clear inherited
CLAUDECODEenv var from CLI subprocess to prevent nested session interference - CLI stderr noise: Filter minified JS stack traces into a single clean error line (
DEBUG=1for full output)
Changed
- claude-agent-sdk: Bumped minimum version to 0.1.48
- Agent mode: No longer prompts for URL (agent navigates autonomously)
- Header UI: Version and task labels now use mode-specific colors
Upgrade
pip install --upgrade reverse-api-engineer
# or
uv tool install --upgrade reverse-api-engineerCloses #51
v0.4.3
Fixed
- AskUserQuestion interactive prompt: Fixed interactive questionary UI not rendering in engineer and agent modes. Switched from
ClaudeSDKClienttoquery()function, asClaudeSDKClientsilently auto-approvesAskUserQuestionwithout triggering thecan_use_toolcallback
Install / Upgrade
pip install --upgrade reverse-api-engineerv0.4.0 - List CLI subcommand
What's New
list CLI subcommand
Non-interactive command for discovering and querying generated scripts and runs.
# Rich table (compact)
reverse-api-engineer list
# Full table with all columns
reverse-api-engineer list --full
# JSON output for piping/scripting
reverse-api-engineer list --json
# Filter by mode, model, or prompt search
reverse-api-engineer list --mode auto --search "hn" --limit 5Features:
- Rich table output (compact and full modes) and flat JSON array output
- Filter by mode (
--mode), model (--model), prompt substring (--search), and result limit (--limit) - Enriches each run with script directory path, file list, file count, and local copy detection
Full Changelog: v0.3.3...v0.4.0
v0.3.3 - GitHub Copilot SDK Support
What's New
Added
- GitHub Copilot SDK: Third SDK option for reverse engineering using GitHub Copilot
- New
CopilotEngineerfor HAR analysis via Copilot subscription (cost: $0) - Auto mode support with MCP browser integration via Copilot SDK
- Interactive
AskUserQuestiontool support for Copilot sessions - Permission handler and tool use hooks for agent visibility
- Install with:
pip install 'reverse-api-engineer[copilot]'
- New
- Comprehensive Test Suite: 593 tests achieving 97.4% code coverage
Changed
- Claude Model Updates: Updated from Claude 4.5 to Claude 4.6 (Opus and Sonnet)
- Sync Filtering: Unified sync filtering logic with relative paths; excludes
node_modules - TUI Improvements: Reduced thinking truncation for better agent visibility
Fixed
- Python 3.11 Compatibility: Fixed multi-line f-string expressions that required Python 3.12+
- Thread Safety: Used
loop.call_soon_threadsafefor Copilot SDK event callbacks - Session Timeouts: Added 10-minute timeout protection for Copilot session completion
- Resource Cleanup: Ensured
CopilotClientis always stopped viatry/finally - HAR Recording: Changed HAR recording content mode to embed and optimized file saving
Install/upgrade:
pip install --upgrade reverse-api-engineer