Migrate the MCP server extension to the shared esp-pylib library:
- replace raw stderr prints with esp_pylib.logger.log (note/err)
- route informational logs to stderr via set_info_stream so stdout
stays reserved for the MCP JSON-RPC transport
- enumerate connected devices via esp_pylib.serial_ports.get_port_names
- stub the new imports in the mcp_ext unit tests
Co-authored-by: Cursor <cursoragent@cursor.com>
Add a `monitor_device` MCP tool that lets an AI agent run a scripted,
non-interactive `esp-idf-monitor` session against a flashed device and
get back a short status plus a log file path, instead of raw serial
output inline.
Under the hood:
- The agent supplies a plain-text command body (expect/send/sleep/reset/
exit/comments). `assemble_monitor_script_from_agent_commands()` frames
it into a script the monitor's non-interactive command mode can
consume via stdin: it appends `exit` if the agent didn't already end
with one, and rewrites every bare `expect <regex>` into `expect
--timeout <timeout_sec> <regex>` via `_monitor_normalize_expect_line()`
(an already-bounded `expect --timeout ...` line is left untouched so
the monitor itself reports a bad value). A leading `reset` is not
prepended - the monitor already resets the chip when it opens the
port - and any `reset` the agent wrote is left in place.
`_monitor_parse_sleep_duration()` extracts each `sleep <n>` duration.
The effective timeout is the sum of every bounded expect duration
plus every sleep duration. Scripts whose sum exceeds
`MONITOR_MAX_SCRIPT_SEC` are rejected. If the script has neither
expect nor sleep (for example only `send`), `timeout_sec` is used so
the process still has a kill bound.
- `monitor_device()` runs `python -m esp_idf_monitor` via
`subprocess.run(..., input=script, timeout=2 * effective_timeout)`.
`no_reset` is forwarded as `--no-reset` so the connection reset can be
skipped; an explicit `-p` is forwarded when a port is given. Extra
arguments match `idf.py monitor` where a build exists: baud (`baud`
tool arg, else `monitor_baud` from `project_description.json`),
toolchain prefix, `--target`/`--revision`, coredump/panic decode, and
ELF files with the app ELF first. The 2x hard timeout is a safety net
independent of the script's own `expect --timeout`/`exit` logic; on
`TimeoutExpired` the process is killed but any output already captured
is preserved and logged. `decode_stream()` normalizes that captured
output, which can be `bytes` on the timeout path even though the
process otherwise runs in text mode.
- The monitor's exit code drives the reported status via
`_monitor_status()`, using `EXIT_EXPECT_TIMEOUT` and
`EXIT_SCRIPT_ERROR` from `esp_idf_monitor.base.constants`: 0 is
success, 110 means an `expect` pattern never showed up before its
`--timeout` elapsed, 2 means the monitor rejected the script (bad
syntax/timeout/regex), anything else is reported generically.
- Serial output and the monitor's own messages share one pipe
(`stderr=STDOUT`) so decoded panic backtraces stay next to the lines
that triggered them. `_save_monitor_output()` writes the full merge to
`<tempdir>/esp_idf_mcp_log/action_monitor/monitor_<timestamp>.log` and
reports a dedicated `Log file:` line. On non-zero exit or process
kill, a short tail of that same merge is also returned inline so the
agent has some failure context without a second file read. If the log
file can't be written, it falls back to inlining a truncated tail.
Closes https://github.com/espressif/esp-idf/issues/18757
Closes https://github.com/espressif/esp-idf/pull/18385
Co-authored-by: Cursor <cursoragent@cursor.com>
Auto-detect used to pick the first Espressif device, even when it did
not match IDF_TARGET. With several boards attached, flash/monitor could
talk to the wrong chip.
Pass the project target into esptool so unmatched ports are skipped.
Resolve the port after ensure_build_directory() so the target is known.
For monitor on an unconfigured project, probe the connected chip and
pass it directly to idf_monitor without configuring the project.
Reading the BUILD_COMPONENTS build property aborts the build with a bare
message stating that the property is unsupported, without naming a
replacement. Add a hint that points at the $<TARGET_EXISTS:idf::component>
generator expression and at the documentation section describing the
migration, so the guidance lives with the other build hints instead of
being spelled out in the build system sources.
FastMCP was renamed to MCPServer in mcp 2.0, which made idf.py mcp-server
report a false "MCP dependencies not available" error. Prefer MCPServer
and keep a FastMCP fallback for mcp 1.x.
Closes https://github.com/espressif/esp-idf/issues/18904
Co-authored-by: Cursor <cursoragent@cursor.com>
Allow specifying extra compiler flags (warning flags, preprocessor
definitions, etc.) via a new 'compile_options' entry in the
idf-build-file frontmatter. The options are applied to the source
file via target_compile_options in the generated main component
CMakeLists, with CMake quoting so that flags like -DMSG="hello world"
reach the compiler exactly as written.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Add a new idf.py extension that allows building standalone C files
without requiring full project boilerplate. Source files can include
optional YAML frontmatter in block comments to specify sdkconfig
options, component dependencies, and target configuration.
When the frontmatter configuration changes, the stale sdkconfig is
removed so the new defaults are applied, and a target change also
clears the build directory since IDF_TARGET is pinned in the CMake
cache.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Migrate idf.py and its actions to the shared esp-pylib library:
- add esp-pylib (and esp-pylib[ide]) to the core/ide requirements
- base FatalError on esp_pylib.errors.FatalError, keeping the idf.py ctx cleanup
- replace the local raw-ANSI helpers (red_print/yellow_print/print_warning/
color_print) with esp_pylib.logger.log (warn/err/note/hint) across the
idf.py actions
- install esp_pylib exception reporting at the idf.py entry point and silence
the logger during shell completion
- update affected tests for the new prefixes/line wrapping
- add normalize_output() helper and unify terminal env in conftest to handle
Rich line-wrapping in CI assertions
Co-authored-by: Cursor <cursoragent@cursor.com>
- esp_tls_mbedtls: require cert when PSA-backed server/client key is set
- esp_tls_mbedtls: drop redundant pk_init/x509_crt_init (calloc handles it)
- psa SE driver: copy callbacks/opaque_key by value (no lifetime coupling)
- psa SE driver: replace atomic CAS with simple null check on register
- psa SE driver: use sig_len from sign callback with bounds validation
- psa SE driver: validate pubkey_len returned by export_pubkey callback
- psa SE driver: check hash sub-alg in RSA PKCS1V15 branch of validate_request
- psa SE driver: align secure_element_register_callbacks doc with value-copy impl
- esp_https_server: initialize server_key in HTTPD_SSL_CONFIG_DEFAULT
- mbedtls: move SECURE_ELEMENT_DRIVER_ENABLED to esp_config.h for parity
with ESP_ECDSA_DRIVER_ENABLED; drop target_compile_definitions
- docs: fix esp_tls_cfg_t -> esp_http_client_config_t cross-reference
- docs: check psa_import_key() status in ESP-TLS PSA example
- hints/error_output: point at CONFIG_MBEDTLS_SECURE_ELEMENT_DRIVER_ENABLED
Any idf.py invocation could hang indefinitely with no output while
spawning an unbounded chain of "idf.py --version" subprocesses,
eventually exhausting system memory.
During init_cli(), idf.py parses the project's dependencies.lock to
vet trusted component-provided idf_ext.py extensions. If the lock
contains a component whose manifest has an "if: idf_version" clause,
evaluating it calls idf-component-manager's _get_idf_version(). Outside
a CMake build the IDF_VERSION environment variable is not set, so that
function falls back to running "idf.py --version" as a subprocess,
which re-enters init_cli() and recurses without bound.
During a normal CMake build the component manager runs as a subprocess
that already has IDF_VERSION in its environment (see build/config.env),
so the fallback is never reached. The recursion happens only because
idf.py runs component-manager code in-process during its own CLI
startup, outside that context.
Seed IDF_VERSION into os.environ early in init_cli(), before any
dependencies.lock parsing, using the subprocess-free
idf_version_from_cmake() helper. This gives in-process component-manager
code the same IDF_VERSION a CMake build would provide.
This feature is useful for 3rd-party software to run GDB with predefined
options that described in project_description.json file
allow to pass custom options to "idf.py gdb":
--gdb-commands: command line arguments for gdb. (without changes)
-ex: pass command to gdb.
-x: pass gdbinit file to gdb. Alias for old --gdbinit command
Previously the error message was
CMake Error at /home/user/esp-idf/tools/cmake/build.cmake:296 (__component_get_property):
__component_get_property Function invoked with incorrect arguments for
function named: __component_get_property
Call Stack (most recent call first):
/home/user/esp-idf/tools/cmake/build.cmake:341 (__build_resolve_and_add_req)
/home/user/esp-idf/tools/cmake/build.cmake:638 (__build_expand_requirements)
/home/user/esp-idf/tools/cmake/project.cmake:710 (idf_build_process)
CMakeLists.txt:6 (project)
Now it will be:
CMake Error at /home/user/esp-idf/tools/cmake/build.cmake:298 (message):
Failed to resolve component 'whatever' required by component 'main'.
Call Stack (most recent call first):
/home/user/esp-idf/tools/cmake/build.cmake:345 (__build_resolve_and_add_req)
/home/user/esp-idf/tools/cmake/build.cmake:642 (__build_expand_requirements)
/home/user/esp-idf/tools/cmake/project.cmake:710 (idf_build_process)
CMakeLists.txt:6 (project)
Also improved the hint to show the component name in quotes.
There is currently a bug in the __build_resolve_and_add_req function in
tools/cmake/build.cmake where the check for registered component
requirements is incorrectly applied to the component itself rather than
its dependencies. This issue likely originated from a typo, using
component_target instead of _component_target. To prevent further
confusion, _component_target has been renamed to _req_target.
Fixing this revealed multiple incorrect dependencies for the Linux
target, which have now been resolved by explicitly specifying the
dependencies for the Linux target.
Closes https://github.com/espressif/esp-idf/issues/13447
Signed-off-by: Frantisek Hrbata <frantisek.hrbata@espressif.com>
Currently we silently ignore when the original component is not found
in a hope we can provide at least some meaningful hint. As it turned
out it's not true. Instead of providing misleading hint, just return
error. This adds several checks for situations, which should not happen,
but when they do it should be easier to identify the root cause of the
problem.
For example when hint module received malformed output with extra new
lines, e.g. caused by a bug in RunTool, it wrongly reported the original
component as source component.
This should also fix the tests on Windows.
Signed-off-by: Frantisek Hrbata <frantisek.hrbata@espressif.com>
