The pipeline wrote to 'Changelog.md' and git-added the same literal;
on Windows the write lands in a CHANGELOG.md (case-insensitive fs) but
git add under the wrong casing does not update the tracked index entry,
so the commit carried a STALE changelog and Lib001's pre-commit
version-match hook rejected the release (v0.22.0.0, 2026-07-24).
resolveChangelogName() picks the real filename for both the write and
the git add.
ScriptIecLanguageObjectContainer.create_pou raises 'out of the range of
valid values / Parameter name: return_type' when creating a Function
without one. Parse 'FUNCTION Name : TYPE' from the provided declaration
(fallback BOOL; the declaration replace overwrites the header anyway).
ScriptPouObjectCollection.remove(name) can return without effect on
SP21 (no exception, entry persists -- observed after a program rename
left a stale task call). Delete by index first and fail loud if the
name is still in the call list after save.
Project close is refused while logged in, which silently defeated the
editor-view flush for entire online sessions -- views accumulated until
the IDE died with 'running low on system resources' (Sea Leopard
2026-07-24, ~30 scripted edits + downloads in one online day). Attempt
online_application.logout() before close; the next online tool call
re-logs-in via ensure_online_connection with pre-registered credentials.
Auto-increment per bump burned through Drive sequence numbers for
builds that never left the machine (KK: 'stay on 006 until further
notice'). The number now advances only when a build is actually
uploaded -- set manually in the GVL at upload time. Seeding when
absent is unchanged.
New GVL variable carrying the Drive export name for the build
(KK convention: '<project stem>_NNN'). Incremented from the previous
GVL value on every bump (zero-padding preserved, manual seeding
respected); seeds '<stem>_001' when absent. Gaps in the sequence mean
a version was never uploaded. Lets the running PLC report exactly
which Drive file it came from.
list_device_parameters and set_device_parameter only iterated
device.device_parameters and connector.parameters. Host-side parameter
sets -- e.g. the WAGO 750-series 'K-BUS Parameters' grid (program start
interlock, k-bus cycle time, TCM control) -- are exposed exclusively via
ScriptConnector.host_parameters (SP21 stub ScriptDeviceParameters.pyi;
helpme-codesys ScriptEngine, ScriptConnector.host_parameters), so both
tools reported zero parameters on the Kbus master node.
Verified on the Sea Leopard PFC200 project: id 150 'program start
interlock' now lists and sets offline.
BUG 1: the README.md version replace was content.replace(/v\d+\.\d+\.\d+\.\d+/g,
newVersion) -- a blanket sweep that rewrote every version-looking string in
the file, corrupting historical fix landmarks, build-archive snapshots, and
even a different project's synced-library version (observed live on
SeaLeopard's README v1.3.0.0 -> v1.3.1.0). Replaced with updateReadmeVersion:
only touches the version token on the first "# " title heading, or a single
"**Version:**" line (Lib001 convention). If no anchor is found (or the
Version-line anchor is ambiguous), nothing is changed and the reason is
reported -- no global-replace fallback.
BUG 2: appendChangelogEntry's "ownership guard" checked for the literal text
'Auto-generated by `bump_project_version`', but the intro this same tool
seeds into new files says 'Auto-appended by `bump_project_version` on
release.' -- the strings never matched, so every one of this tool's own
previously-created Changelog.md files (SeaLeopard, since its v1.0.0.0 seed)
was treated as foreign and silently skipped. The skip only logged to
stderr while release_project_version's caller printed a fixed "Changelog.md:
appended vX" success line regardless. Separately, a genuinely hand-maintained
Keep-a-Changelog file (Lib001's CHANGELOG.md, "## [Unreleased]" / "## [x.y.z.w]
- date") was never understood at all. Confirmed via SeaLeopard commits
3c87d18 / c98c264, which hand-repaired both the missing changelog entries and
the corrupted README.
Replaced with buildChangelogUpdate (pure) + appendChangelogEntry (I/O
wrapper): detects the existing file's heading style from its own headings
(not intro wording) and emits a matching entry -- "ours" style
(## vX.Y.Z.W -- date) inserted before the first existing entry, or
Keep-a-Changelog style (## [X.Y.Z.W] - date) inserted after any
"## [Unreleased]" section (or before the first version heading if none).
An unrecognized format is left untouched with a clear skip reason. The write
is verified by re-reading the file before release_project_version reports
success; a write that didn't happen is now reported as NOT appended.
Rebuilt dist/ (tsc + scripts copy) since the MCP server runs from there.
Tests: 16 new (update-readme-version, build-changelog-update,
append-changelog-entry), full suite: npx vitest --run -- 229 passed, 1
pre-existing unrelated failure (script-manager.test.ts trips on a local
__pycache__ dir under src/scripts, present before this change too).
Per feedback: emit '<Namespace> : STRING := "<version>";' (self-describing,
name carries the library) instead of sLibNN with the name inside the value.
Namespace sanitized to a valid IEC identifier; deduped by name; sVersion /
uiLibraryCount reserved-name guarded.
Per feedback: emit one 'sLibNN : STRING := "Namespace Version"' per reference
(same style as sVersion, individually online-readable) instead of
asLibraries ARRAY[1..N] OF STRING(79). uiLibraryCount unchanged.
Enumerate all library-manager references (name + resolved version) and write
them into the runtime-readable GVL as uiLibraryCount + asLibraries ARRAY OF
STRING(79), alongside sVersion. Refreshed on every version bump so the running
PLC reports its full library manifest. Best-effort enumeration (soft-fails to
an empty manifest so a bump never fails); ARRAY omitted when empty. Enumeration
reuses the verified list_project_libraries API (has_library_manager -> lm.references).
Not yet active in a running MCP: needs a server restart (+ rebuild/republish if
scripts are bundled) and a test bump to verify manifest population and that
read_running_version_online still parses sVersion.
project.import_native lands at the PROJECT ROOT (POU pool, visible only in
the POUs view), and CODESYS refuses to move root-level objects into an
application afterwards ('Cannot move X from <root>') -- a wrong-level bulk
import is unrecoverable by script. ScriptObject.import_native (API 3.4.4.0)
imports under a node; expose it via parentObjectPath so library subtrees
can be imported straight into e.g. 'Application/MRLib'.
CODESYS persists the open-editor window layout in
<Project>-<user>-<machine>.opt and restores it on project open -- so a
plain close/reopen (or even a full IDE restart) brings every accumulated
editor view straight back, as seen live: a fresh IDE hit 'low on system
resources' immediately after reopening a project whose .opt held ~60
views (647 kB sidecar). Deleting the per-user .opt (pure UI state,
regenerates clean; AllUsers.opt kept) makes the flush actually stick.
Every scripted textual_declaration/textual_implementation write and object
creation opens an editor view in the visible IDE. The ScriptEngine has no
API to close views (ScriptCommands is lookup-only per ScriptSystem.pyi and
helpme-codesys ScriptingEngine docs; the WinForms menus are also invisible
to UI Automation). After ~40-60 scripted edits the IDE exhausts UI
resources ('Please close some views to free up resources') and every
subsequent script call times out at the IPC layer.
Mitigation: count edit-tool calls (set_pou_code, create_pou/method/
property/dut/gvl); before the Nth edit since the last flush, run
flush_editor_views.py which saves, closes and reopens the primary project
-- disposing all editor views in seconds. Threshold configurable via
CODESYS_EDITOR_FLUSH_THRESHOLD (default 20, 0 disables). The flush result
is surfaced as a NOTE/WARN line in the tool response.
Both tools silently DISCARDED declarationCode/implementationCode (zod strips
unknown keys): the object was created EMPTY, compiled clean, and did nothing.
Cost a full download/debug cycle on the fp-j1939 bench session 2026-07-17.
Now the params are real: applied after creation via ScriptTextualObject
textual_declaration/textual_implementation.replace() (same proven API as
set_pou_code), failing LOUDLY if provided code cannot be applied.
Also extend findReservedIecIdentifiers beyond time-suffix letters to the full
IEC 61131-3 ST reserved-keyword set, checked case-insensitively ('by'/'BY' are
both the FOR-loop step keyword -- 'by : BYTE;' failed on the bench 2026-07-16).
Sources: IEC 61131-3 (3rd ed.) keyword tables; CODESYS export-format keywords
(content.helpme-codesys.com/en/CODESYS%20Development%20System/_cds_keywords.html).
Standard-function names (MIN/MAX/ABS/...) deliberately excluded: not confirmed
compiler-rejected, and a false positive blocks legitimate code. Scanner now also
checks every name in comma-separated declaration lists and AT %address forms.
Guard wired into create_pou + create_method declarationCode, same as set_pou_code.
Tests: tests/unit/reserved-iec-identifiers.test.ts (8 cases); full suite 214 green.
set_gateway_and_address raises 'Invalid address format' for 'ip[:port]'
strings. Detect IP form and route it to ScriptDeviceObject
.set_gateway_and_ip_address (helpme-codesys ScriptDeviceObject, since
3.5.8.0), which binds the block driver directly by IP -- needed for
SSH-tunnelled PLCs (e.g. 127.0.0.1:11740 -> lab PLC via jump host).
Live-verified against WAGO PFC200 10.0.0.202 through codesys1-prox.
Two field-observed defects in bump_project_version (hit on nmea2000-mr-library):
1. appendChangelogEntry wrote into 'Changelog.md', which on a case-insensitive
Windows filesystem is the SAME file as a repo's hand-maintained
CHANGELOG.md. With Keep-a-Changelog '## [x.y.z.w]' headings (no '## v'),
the appender fell through to the append-at-end branch and dumped an auto
entry at the bottom of the curated file. Now: if the existing file was not
created by this tool (missing the auto-generated intro marker), skip the
append with a warning instead of writing.
2. First-run seed ignored existing releases: with Project Information.Version
unset but the repo tagged up to v0.8.0.0, a manual minor bump seeded
1.0.0.0 -- out of series. Now the TS side derives a seed from the latest
v* tag bumped at the requested level (new SEED_VERSION script param);
the Python side falls back to the classic 1.0.0.0 only when no usable
tag exists.
tsc clean, vitest 206/206.
Live verification against the PFC200: device-level download_source on
SP21 writes its temp file into the CODESYS install dir (Program Files)
and dies with access-denied; the failed attempt leaves a locked
Archive.prj in the project dir that breaks the app-level fallback with
'file is being used by another process'. Now: remove stale Archive.prj
up front, use app-level source_download unless compact was explicitly
requested.
28 of the 46 new tools verified against CODESYS V3.5 SP21 Patch 5; 7 SP21
stub-vs-runtime drift bugs found live and fixed. Remaining live checks:
import_text_list_file and the 12 online tools (need the WAGO PFC200).
SP21 removed IScriptUser.change_password ('no longer supported'). The
script now probes change_password/set_password/reset_password and, if
none works, saves the user WITH A WARNING instead of failing the whole
creation. Creation is idempotent so a half-failed prior run doesn't
block a retry on the duplicate name.
Same overload drift as export_xml: positional (path, None, folders) put
the path string in the reporter slot ('expected IImportReporter, got
str'). All args now passed by keyword with positional fallback.
Live verification on SP21: users.create raised "permission 'Modify' not
granted to user '(nobody)'". The scripts now log in as 'Owner' with empty
password (CODESYS default) when nobody is logged on; new optional
adminUser/adminPassword args override for protected projects.
Live verification: positional (objects, None, path, ...) bound 'objects'
into the reporter slot on SP21's overload, silently switching export_xml
into export-to-string mode -- tool reported success with no file written.
All args now passed by keyword; export_plcopen_xml and export_native
raise if the destination file does not exist after the call.
Live verification against CODESYS V3.5 SP21 Patch 5: calling it raised
'bool is not callable'. Handle both the property (SP21) and callable
(stub-documented) shapes.
1. find_object_by_path: final name verification used the original
full_path instead of the dot->slash normalized segments, so
dot-separated paths ('Application.MyPOU') traversed correctly but
failed the final check and returned None.
2. User-arbitrary values (plcPath/plcDirectory, passwords, comments,
project-info fields, device parameter name/value, task event,
device credentials) were interpolated into r"..."/r"""...""" Python
literals unescaped -- a quote or triple-quote in the value broke the
generated script (or injected code). Templates now take pre-escaped
literals via pyStringLiteral().
Reviewed-range: dead49a..e9aa714. Third reviewer finding (task.priority
must be int) was rejected: SP21 ScriptTaskConfigObject.pyi types the
priority/interval/interval_unit setters as str.
- add_library.py line 49 had UTF-8 box-drawing dashes in a comment with no
coding declaration -- latent IronPython 2.7 source-encoding risk.
- build now removes dist/scripts before copying, so deleted/renamed
templates (compile_project.py.bak, probe_app_error_state.py,
set_library_namespace.py) no longer ship in the npm tarball.
- script-manager test now asserts EVERY template is ASCII-only instead of
per-phase lists.
102 tools total. SP21 ScriptEngine functional coverage complete per plan
(SVN/Composer/Cloud deliberately out of scope; plug_module and trace
tools deferred).
String.replace with a string replacement interprets $$/$& as regex
replacement patterns, corrupting IEC string literals like '$R$N' passed
through tool params (set_pou_code code bodies, write_variable values).
Use a function replacement so values pass through verbatim.
Persistent mode now uses a LazyPersistentExecutor everywhere the launcher
is not running: --no-auto-launch before first use, after shutdown_codesys,
and after a launch conflict. The first tool call launches the VISIBLE IDE
and delegates; nothing silently spawns --noUI processes anymore.
Why: headless spawns pop modal dialogs nobody can see (tool calls just
abort), hold .project locks (stale .~u files), and leave orphaned
CODESYS.exe processes that then block the next launch. Diagnosed live on
2026-06-11 while deploying TestN2k_v2_Fable: open_project/download ran
headless via the --no-auto-launch fallback and burned ~30 min on invisible
dialogs, zombie PIDs and lock-file cleanup.
Headless execution now requires explicit opt-in: --mode headless or
--fallback-headless.
Also lands the task-configuration tools from the 2026-06-09 session that
were complete but uncommitted: list_tasks, add_pou_to_task,
remove_pou_from_task (+ their IronPython scripts).
The server is a standard stdio MCP server and works with any MCP client, not
just Claude Code. Added a "Use with OpenAI Codex CLI" section covering the
~/.codex/config.toml location, the [mcp_servers.<name>] TOML table (the
equivalent of the --print-config JSON block), env-based credential injection,
tool_timeout_sec for slow CODESYS launches, and the codex mcp add shortcut.
Linked it from Quick Start step 3.
registerIdeBridgeTools() spawned CodesysMCPBridge.exe but the SIGINT/SIGTERM shutdown handler never closed the client, so every orchestrator exit left the shim running. They piled up across sessions (7 observed live with no IDE open).
- ide-bridge: add findOrphanedBridgePids()/killOrphanedBridges() using dead-parent detection (only reaps shims whose parent process is gone, so live sessions are untouched); harden close() with a taskkill /F fallback; expose .pid getter.
- server: sweep orphaned shims at startup, track the bridge client and close() it on shutdown, and add a process.on('exit') taskkill safety net for non-signal exits (stdin EOF / fatal).
Verified end-to-end against compiled dist: real orphan detected+reaped, live-parent shim left alone. Build clean, 156 tests pass. Bump 0.10.1 -> 0.10.2.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
CODESYS opens/saves projects from UNC paths (\server\share\...) only
unreliably -- it tends to fail late and opaquely. Add src/path-guard.ts
(isUncPath + uncPathError) and gate open_project, create_project,
save_project, and launch_codesys_with_project on it: each now returns
isError early telling the user to map a drive (net use Z: \server\share)
or copy the project to a local drive. Mapped drive letters and local
drives (incl. \?\C:\) are not treated as UNC, so existing workflows are
unaffected.
Also bumps version to 0.10.1 (first published release of the 0.10.x line;
includes the prior unreleased phobiCS-tui removal and add_device tool).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Pairs with the add_device MCP tool that landed in e1f47a1 (folded in
during the approve-gate rip). Wraps ScriptDeviceObject.add(name, device_id)
to programmatically attach a child device under a parent device:
'Modbus TCP Server' under an Ethernet adapter, 'Ethernet' under the top-
level PLC, etc.
Behaviour:
- Required inputs: PROJECT_FILE_PATH, PARENT_PATH ('/'-separated, e.g.
'MainPLC' or 'MainPLC/Ethernet1'), DEVICE_NAME, TARGET_NAME (substring
of device repository display name).
- Optional: TARGET_VERSION (exact, e.g. '4.5.0.0'). Omit -> highest-version
match wins, mirroring update_device_type's resolver.
- Suppresses CODESYS prompt dialogs via PromptHandling.NONE (with int=0
fallback for SP21).
- Idempotent: if a child with DEVICE_NAME already exists under the parent,
no-ops with SCRIPT_SUCCESS rather than creating a duplicate or erroring.
- Refuses when:
- parentPath isn't found in the project,
- the object at parentPath isn't a device (only attaches under device-
typed ScriptObjects),
- TARGET_NAME (+ optional version) has no match in the device repo --
error suggests inspecting Tools > Device Repository.
- On ScriptDeviceObject.add() failure prints the full traceback plus the
two common causes (parent doesn't allow this child type; device
descriptor needs an uninstalled library).
After add, saves the project. Save failure is a WARN, not a hard error.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Knock-on cleanup from the phobiCS-tui nuke in 4b82b7c. With the TUI gone
the gate had no UI to render -- it just printed a stderr warning and
auto-approved on every call. Ripping it out kills ~470 lines of dead code
and a redundant CLI flag.
Removed:
- src/approve-gate.ts (whole file; runApproveGate + gateOpForTool + helpers)
- tests/unit/approve-gate.test.ts
- bin.ts: --approve-edits CLI option, approveEdits config wiring, startup log
- types.ts: ServerConfig.approveEdits field + its doc comment
- server.ts: import of runApproveGate/gateOpForTool, the `Approve edits:` log,
16 `gateOpForTool({...}); if (blocked) return blocked;` blocks across
every modifying MCP tool handler, and the lone `runApproveGate({...})`
block in set_pou_code.
Folded in (Karstein's prior WIP, gate-free now per his explicit choice
"Single rip commit, I edit your WIP too"):
- New `add_device` MCP tool in server.ts that wraps ScriptDeviceObject.add
for attaching child devices (Modbus TCP Server under Ethernet, Ethernet
under PLC, etc.). The supporting `src/scripts/add_device.py` remains
UNTRACKED in Karstein's working tree -- still his to commit separately.
Without that script the tool will fail at runtime; with it, fine.
Verify:
- `npx tsc --noEmit` clean
- `npm run build` clean
- `npx vitest --run tests/unit/` -> 14 files, 119 tests, all pass
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The phobiCS-tui CLI/UI is retired in favour of driving the codesys-mcp-sp21-plus
MCP tools directly from Claude Code (see https://docs.anthropic.com/claude-code).
The MCP server itself is unchanged; only the TUI front-end goes.
Removed:
- src/tui/ (browser + approve + shared + entry; 14 files)
- dist/tui/ (compiled output)
- tests/tui/ (14 .test.* files + the mini-mirror fixture tree)
- tsconfig.tui.json
- 3 superpowers plans/specs docs (2026-04-28 phobics-tui v0.1-v0.2, 2026-04-29 v0.3-live-values, 2026-04-28 tui-design)
- package.json: phobiCS-tui bin entry, build:tui script, TUI compile step in build, TUI typecheck step
- package.json: dependencies ink + react + diff (TUI-only); devDependencies @types/diff + @types/react + ink-testing-library
- README.md: ## phobiCS-tui section + ### Inline live values subsection
Git history side:
- Worktree .worktrees/phobics-tui removed (was on feature/phobics-tui-followup @ 7e427e9)
- Local + origin branches deleted:
- feature/phobics-tui (was 08ee361, 0 unmerged vs origin/main)
- feature/phobics-tui-followup (was 7e427e9, 0 unmerged vs origin/main)
- feature/phobics-tui-v0.3-live-values (was 9f4dc48, 0 unmerged vs origin/main)
- All three branches were merged into main, so deleting refs loses no history --
the commits remain reachable through main.
Knock-on (deliberately deferred):
- src/approve-gate.ts and the --approve-edits flag in src/bin.ts / src/server.ts
still exist. With the TUI gone, the gate auto-approves at every prompt and
prints a `[approve-gate] No TTY available -- phobiCS-tui cannot render` warning
to stderr. The infrastructure also still has callers in the uncommitted
add_device work, so a clean rip-out is left for a follow-up.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Once launch() refused due to a foreign CODESYS, the error string was
cached in lastError and getStatus() returned that frozen snapshot
forever. Closing the foreign CODESYS did not update the status -- only
an MCP restart cleared it. From the user's POV: "every time CODESYS
HAS been open, you get a problem".
Fix: getStatus() now calls revalidateLaunchRefusal() first. If the
launcher is parked in 'error' state with a "Refusing to launch:"
prefix and findConflictingInstances() now returns empty, we transition
back to 'stopped' and clear lastError so the next status call / launch
attempt sees a fresh state. Other 'error' states (process died,
watcher timeout, exe not found) are not auto-cleared -- only the
launch-refusal cache, since that's the one that goes stale on its own
when the user closes the foreign window.
Bump 0.9.13 -> 0.9.14.
Tests: 2 new launcher unit tests pin the auto-clear behaviour and
verify unrelated 'error' states are NOT auto-cleared. 22/22 launcher
tests pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
