signata-swe
AUTOSAR & non-AUTOSAR SWE.3 AI skills (project-init, code-dev, code-fix, code-review, unit-test) plus DOORS SWE.5/SWE.6 test-case generation (integration-test, qualification-test).
Plugin signata-swe from marketplace signata in Signata-GmbH/signata-swe-claude-skills. Commands read from the repo's manifest — 13 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
What it installs
Skills (7)
code-devImplement an application module (or a feature in an existing module) from a filtered set of software requirements, following the project's conventions. Adapts automatically to the repo's project.type (AUTOSAR: RTE/runnables/MemMap; non-AUTOSAR: _cfg.h IF-macros/scheduler). Two-phase workflow with a
code-fixAnalyse a reported defect in an existing module and produce a root-cause analysis plus a minimal, traceable fix — from test-bench evidence (debugger dumps, CAN/LIN traces, DTCs, a failing SWE.4/5/6 case, scope measurements), a pointer to the suspect code or requirement, or a code-dev/code-review fol
code-reviewPerform a formal SWE.3 software code review of a module's user-implemented C, producing a populated Findings List. Adapts automatically to the repo's project.type (AUTOSAR: MISRA rule-numbers in scope, AUTOSAR-isms expected, Critical severity; non-AUTOSAR: MISRA out of scope, no AUTOSAR-isms, four s
integration-testGenerate draft software integration test cases (ASPICE SWE.5) for one module's interfaces — each case names both ends, the module and the peer at the other end of the port — as a 3-sheet Excel workbook (Test Cases / Traceability / Open Points) an engineer reviews and enters into DOORS by hand. Test
project-initCreate (or validate) the per-repository AI project config 20_AI/ai_project.yaml — once per repo, on the repo's base branch, so every engineer and every module shares one config instead of conflicting copies. Detects AUTOSAR vs non-AUTOSAR, discovers layout roots and input documents, asks only for wh
qualification-testGenerate draft software qualification test cases (ASPICE SWE.6) for one feature, as a 3-sheet Excel workbook (Test Cases / Traceability / Open Points) an engineer reviews and enters into DOORS by hand. Runs in the SWE.3 project repository like the code skills, with its own test-spec config (20_AI/ai
unit-testGenerate or update a VectorCAST unit-test (.tst) suite for a target module from its requirements and code. Adapts automatically to the repo's project.type (AUTOSAR or non-AUTOSAR). Authors by inspection only — never builds, runs, or reports coverage it did not measure. Use when asked to create, exte
Hooks (4)
- SessionStart
python "${CLAUDE_PLUGIN_ROOT}/hooks/run_tracker.py" session-start
- UserPromptSubmit
python "${CLAUDE_PLUGIN_ROOT}/hooks/run_tracker.py" prompt
- PostToolUse
python "${CLAUDE_PLUGIN_ROOT}/hooks/run_tracker.py" skill
- Stop
python "${CLAUDE_PLUGIN_ROOT}/hooks/run_tracker.py" stop
Maintain signata-swe?
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.