- resolve request model against current available catalog
- choose safe default when configured default id is not present
- add tests and docs for daemon request-model fallback behavior
- add cursor model fields to unified config defaults and YAML comments
- expand Cursor model catalog/defaults and return configured current model
- keep cursor.settings.json env model fields in sync with config updates
- align unit tests for new model defaults and route merge behavior
Refs #555
When abort fires after streaming begins, onAbort closed HTTP/2 handles
but never closed the ReadableStream controller. Consumers waiting on
reader.read() could hang indefinitely since end/error handlers were
gated by streamClosed flag.
Hoist controller reference to outer scope so abort handler can close it.
- Check response.status before calling transformProtobufToSSE in fetch fallback
- Move abort signal registration before response headers to handle early cancellation
- Route all `ccs cursor *` to handleCursorCommand (no profile-switching)
- Add --ccs-daemon marker to spawned process for stable PID validation
- Use random port in lifecycle integration test to prevent CI conflicts
- Remove unnecessary .toFixed(1) on integer tokenAge
- Validate PID belongs to cursor daemon via /proc before signaling
- Tighten detectProvider regex to avoid over-matching o-prefixed models
- Add integration test for daemon start→health→stop lifecycle
- Add void cast on discarded handleHelp() return value
- Update model catalog date comment
- Export detectProvider and formatModelName from barrel index
- Export CURSOR_SUBCOMMANDS from cursor-command.ts, import in ccs.ts
- Use getCcsDir() in tests instead of manual path construction
- Add handleCursorCommand routing test for unknown subcommand
- Add removePidFile() in safeResolve on failure to prevent stale PIDs
- Only send SIGKILL in stopDaemon if SIGTERM wait loop exhausted
- Check isDaemonRunning before model fetch to avoid 5s timeout
- Add port validation unit tests for startDaemon
- Change spawn stdio from piped to 'ignore' preventing buffer deadlock
- Replace setInterval with sequential setTimeout polling
- Fix TOCTOU in SIGKILL escalation (send directly without probing)
- Move CursorConfig interface to types.ts
- Change detectProvider default from 'openai' to 'unknown'
- Remove redundant removePidFile() when PID is null
MEDIUM fixes:
- Kill orphaned process when health check times out after 30s
- Prevents zombie processes from consuming resources
LOW fixes:
- Return exit code 1 for unknown cursor subcommands (was 0)
- Simplify exit handler dead code branch (else if code !== null → else)
- Add double-resolve guard in startDaemon with safeResolve wrapper
- Add port validation (1-65535) before Node.js script interpolation
- Fix misleading comment in stopDaemon (no PID file handling)
- Add getDaemonStatus test for no daemon running case
- Add stopDaemon tests for graceful non-existent PID handling
- Add SIGKILL escalation in stopDaemon after SIGTERM timeout
- Document router difference between cursor/copilot help behavior
- Fix detectProvider to handle o1/o4 models via regex pattern
- Add fetchModelsFromDaemon fallback test for unreachable daemon
- Update CLAUDE.md help table with cursor command entry
- Wire cursor command into main router (ccs.ts) following copilot pattern
- Centralize default port (4242) and model (gpt-4.1) as constants
- Remove duplicate types already in cursor-protobuf-schema from dev
- Handle daemon exit code 0 before health check succeeds
- Export PID helpers and model utils for testability
- Add unit tests for cursor-daemon (PID file ops, isDaemonRunning)
- Add unit tests for cursor-models (detectProvider, formatModelName, defaults)
Use specific patterns ('rate limit', 'resource_exhausted', 'too many
requests') instead of broad 'rate' or 'limit' substrings that would
false-positive on unrelated errors like 'character limit exceeded'.
HIGH PRIORITY FIXES:
- Extract shared buffer parsing logic into parseProtobufFrames generator
method (DRY violation fix) - both JSON and SSE transformers now use
common frame parsing loop, eliminating ~60% code duplication
- Use COMPRESS_FLAG constants instead of hardcoded 0x01/0x02/0x03 in
parseConnectRPCFrame for better maintainability
MEDIUM PRIORITY FIXES:
- Return empty buffer on gzip decompression failure (prevents silent
data corruption) - ALREADY FIXED
- Add debug warning for unknown message roles in convertMessages
- Create GitHub issue #535 for FIELD namespace refactoring follow-up
- Add test coverage: transformProtobufToSSE, error paths, unknown roles
LOW PRIORITY FIXES:
- Rename checksum test to clarify timestamp granularity (~16 min)
- Fix debug log function name in SSE transformer - FIXED BY REFACTOR
- Add comment to TOOL_RESULT_RESULT field documenting future use
All tests pass (1593 pass, 0 fail)
All validation checks pass (typecheck + lint + format + tests)
- Add comprehensive unit tests for cursor-auth.test.ts
- validateToken: valid/invalid tokens, short tokens, UUID formats, empty strings
- extractUserInfo: JWT parsing, email handling, non-JWT tokens, malformed base64
- saveCredentials/loadCredentials: round-trip, invalid JSON/types, missing fields
- checkAuthStatus: authenticated/not authenticated, expired tokens, JWT exp, invalid dates
- deleteCredentials: delete existing/non-existent files, multiple deletes
- All tests use CCS_HOME env var for isolation, real file I/O, no mocks
- Fix dead try-catch around new Date() in checkAuthStatus()
- Replace try-catch with isNaN check (new Date('garbage') returns Invalid Date, not throw)
- Properly handle Invalid Date by checking isNaN(getTime())
- Fix email populated with sub claim in extractUserInfo()
- Change email: decoded.email || decoded.sub to email: decoded.email || undefined
- Prevent non-email values (UUIDs) from populating email field
- Add type guards for JSON.parse result in extractUserInfo()
- Cast to Record<string, unknown> and validate types
- Use typeof checks for email, userId, exp fields
- Enhanced extractUserInfo to return JWT exp claim for expiry detection
- Updated checkAuthStatus to use JWT exp when available, fallback to importedAt heuristic
- Added typeof validation for loadCredentials fields
- Added ENOENT detection for sqlite3 availability in queryStateDb
- Added deleteCredentials function for cleanup
- Updated PR #526 description to remove refreshToken mention
- Fix bitwise shift overflow in generateChecksum: use Math.trunc
division for >>40/>>32 which wrap modulo 32 in JS
- Add FIXED64/FIXED32 bounds checks in protobuf decoder to prevent
out-of-bounds slice on truncated buffers
- Consolidate COMPRESS_FLAG to single definition in schema (DRY):
executor now imports from schema, added GZIP_ALT/GZIP_BOTH values
- Fix token split edge case: use indexOf+slice instead of split('::')[1]
to handle tokens containing multiple :: delimiters
- Fix AbortSignal listener leak: store handler ref, use once:true,
remove listener on request end/error
CRITICAL FIX:
- CursorCredentials interface now matches types.ts (machineId, ghostMode as top-level)
- Fixes runtime error when cursor-auth saves credentials and cursor-executor reads them
HIGH:
- Replace 18+ non-null assertions with guard clauses across executor and decoder
- Prefix unused params in translator (_model, _stream, _credentials)
- HTTP/2 client closes on connection error to prevent leak
- AbortSignal listener leak documented with TODO (inline arrow prevents cleanup)
MEDIUM:
- Export concatArrays from encoder, remove duplicate from protobuf.ts
- Varint decoder now enforces 5-byte max to prevent overflow
- Buffer slice bounds check prevents out-of-range read
- Empty messages array validation with explicit error
- Buffered streaming limitation documented with TODO comment
All edge cases from code review now addressed.
- Convert PID_FILE constant to getPidFilePath() function to respect CCS_HOME changes at runtime
- Add proc.on('exit') handler to clear interval on silent process crashes
- Ensures test isolation by computing paths dynamically
- Split protobuf into 3 files: schema, encoder, decoder for <200 LOC constraint
- Implement ConnectRPC wire format encoding (varint, fields, messages, frames)
- Implement protobuf decoder with gzip decompression support
- Port Cursor executor with HTTP/2 support and checksum generation (Jyh cipher)
- Add OpenAI to Cursor message translator (system→user, tool results handling)
- Transform protobuf responses to both SSE and JSON formats
- Use Node.js built-in crypto/zlib (no uuid dependency)
- All files TypeScript strict mode compliant
Tam Nhu Tran