rds-dv
01Agent Claude Code
Writes cocotb testbenches under val/ for this repo using the RDS-DV framework BFMs. Use for new tests, coverage closure, and reproducing a failing seed. Does not modify RTL.
Agent Claude Code
Writes cocotb testbenches under val/ for this repo using the RDS-DV framework BFMs. Use for new tests, coverage closure, and reproducing a failing seed. Does not modify RTL.
Agent Claude Code
Writes and runs SymbiYosys formal properties under formal/ for this repo. Use for proving contracts, CDC and sizing invariants, and for explaining counterexamples. Treats a vacuous pass as a failure.
Agent Claude Code
Runs regressions for this repo and triages the failures into RTL defect, test defect, flake or infrastructure. Use for suite runs, failure triage and quarantine decisions. Routes findings; does not fix them.
Agent Claude Code
Writes and modifies synthesizable SystemVerilog under rtl/. Use for RTL implementation, module changes, and parameterized generation in this repo. Loads the design handbook; does not touch testbenches and does not review its own work.
Agent Claude Code
Read-only SystemVerilog reviewer for this repo. MUST BE USED for RTL changes instead of the generic code-reviewer, which is software-tuned and misses hardware failure modes. Reports findings and changes nothing.
Settings file Claude Code
Agent settings declaring 64 allowed tools.
Skill Claude CodeCodex
Coverage methodology - Verilator line/toggle coverage per component, functional coverage expectations, and the monbus packet-type coverage matrix. Use when adding tests, closing coverage, or preparing a board-gate sign-off.
Skill Claude CodeCodex
The Sherpa documentation pipeline - mdtodocx with --style, LoF/LoT/LoW caption encoding, book indexes, RTL PDF generation, HAS/MAS vs operator guides. Use before generating or restructuring any deliverable doc.
Skill Claude CodeCodex
What kind of documentation lives where in this repo - handbook (method) vs docs/markdown (reader-facing) vs vault/Tasks (work) vs beside-code (CLAUDE.md, PRD). The rule that a beside-code README is a link, never a standalone guide, and that style guides / how-tos never live in the RTL tree. Use before creating…
Skill Claude CodeCodex
Filelist rules - EVERY module MUST have a .f and MUST be registered in bin/filelists.toml. Components own their compile closure, consumers -f include, never hand-list another area's sources. Use when adding ANY new RTL module, test, or .f file.
Skill Claude CodeCodex
Formal verification flow - SymbiYosys via sv2v flatten, in-RTL ifdef FORMAL properties, mutation-checking every new assertion, harness vacuity traps. Use when writing properties, regenerating proofs, or after changing any module with a formal dir.
Skill Claude CodeCodex
Minimize state machines, and use NONE on the data path. Use when designing or reviewing any block that moves beats, or before writing a state machine of any size.
Skill Claude CodeCodex
HARD RTL design guidelines - reset macros, CDC, valid/ready contracts, streaming no-FSM pattern, SRAM rules, sizing invariants, signal-naming audit. Use when writing or reviewing any RTL.
Skill Claude CodeCodex
Signal-contracts + K-map workbooks (xlsx) for a component - contract sheets per interface and computed Karnaugh maps for key combinational decisions. Use when documenting or reviewing decision logic in engines/schedulers/arbiters.
Skill Claude CodeCodex
Authoring a per-module reference page under docs/markdown (rtl-common/rtl-math/rtl-amba) - the local .md style (header table, Overview, Architecture, Top-level Interface, Data model, Timing, Related, Test) and the SV header block that mirrors it. Use whenever you add or change an RTL block and need its docs/markdown…
Skill Claude CodeCodex
RDS-DV testbenches compose THREE ORTHOGONAL choices - BFM (who drives), sequence (what traffic), randomization (what timing). Pick each deliberately; using the right BFM says nothing about whether the test stresses anything. Points at the authoritative per-family RDS-DV docs. Use when starting or reviewing ANY…
Skill Claude CodeCodex
Use the RDS-DV framework BFMs (GAXI/AXI4/AXIL/APB/AXIS/MonBus) instead of hand-rolling drivers, monitors, or packet decoders. Covers the factory map, decision tree, and the known traps. The BFM is only ONE of three orthogonal axes - see also rds-dv-axes (sequences) and rds-dv-randomization (timing).
Skill Claude CodeCodex
Delay/traffic randomization in RDS-DV - the 19 named FlexConfigGen profiles, which layer to use, and the rule that randomized traffic does NOT prove fairness or arbitration correctness. Use when shaping stimulus timing, writing an arbiter/scheduler/picker test, or choosing a delay profile.
Skill Claude CodeCodex
How to run a regression correctly - always "make clean-all && make run-all-{gate,func,full}-parallel", never a bare pytest. Use whenever running or re-running any test suite in val/ or projects/components/, or before reporting a suite as passing.
Skill Claude CodeCodex
External doc-review rounds (Kimi) - bundle building, serial dispatch, round protection, token budgets, findings triage (doc-fix vs RTL-fix), measuring integration status rather than trusting commit history, verifying fixes with clean rebuilds + mutation checks, and the humanization pass. Use when sending docs for…
Skill Claude CodeCodex
Signal naming - r means flopped, w means combinational. Use when writing, renaming or reviewing any RTL signal, or when counting pipeline latency from a module you did not write.
Skill Claude CodeCodex
Where project tasks/TODOs live and how to track them - the /vault/Tasks/ / directory with open/active/deferred/closed/dropped lifecycle pages. Use whenever you would record a TODO, start/finish a task, or are tempted to create a TASKS.md or TODO.md next to code.
Skill Claude CodeCodex
Auditing test collateral with the review pipeline - what to grab from val/ , bin/TBClasses, and the RTLDesignSherpa-DV framework (golden), the bundle layout, and the audit checklist (three levels, structure, filelists, seeds, BFM usage, real checks). Use when reviewing/auditing tests in an area, or building the…
Skill Claude CodeCodex
The UART characterization-harness methodology - one host program running identically against cocotb sim and the FPGA. Sim transport, regmap-by-name, board bring-up gotchas (JTAG serials, ttyUSB). Use for any board characterization flow.