Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add agodianel/esp32-claude-workbench --skill esp32-arch-reviewgit clone --depth 1 https://github.com/agodianel/esp32-claude-workbenchWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/agodianel/esp32-claude-workbench/esp32-arch-review)<a href="https://agentmods.dev/skills/agodianel/esp32-claude-workbench/esp32-arch-review"><img src="https://agentmods.dev/badge/skills/agodianel/esp32-claude-workbench/esp32-arch-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/agodianel/esp32-claude-workbench/esp32-arch-review"><img src="https://agentmods.dev/badge/skills/agodianel/esp32-claude-workbench/esp32-arch-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00124 | $0.01579 |
| Opus 5.5 | $0.00050 | $0.00632 |
| Sonnet 5.5 | $0.00025 | $0.00316 |
| Haiku 4.5 | $0.00012 | $0.00158 |
Grade A, and why
esp32-arch-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ESP32 Architecture Review
Structured, evidence-based review of embedded-specific correctness. Read-only unless the user asks for fixes (fixes then go through an implementation contract).
Tools
If the esp32-workbench MCP server is connected, call its tools; otherwise run the
CLI. If esp32wb is not on PATH, use
uvx --from git+https://github.com/agodianel/esp32-claude-workbench esp32wb <args>.
Add --json for machine-readable output.
| Need | MCP tool | CLI |
|---|---|---|
| All static checks (pins, stacks, sdkconfig, partitions) | project_doctor |
esp32wb doctor . [--target T] |
| Undersized task stacks | stack_check |
esp32wb stacks . |
| GPIO conflicts for the real chip | pin_audit |
esp32wb pins . |
| Risky / debug sdkconfig options | sdkconfig_review |
esp32wb sdkconfig sdkconfig |
| Partition table sanity | partition_check |
esp32wb partitions partitions.csv --sdkconfig sdkconfig |
| API / peripheral behaviour questions | search_esp_docs |
esp32wb docs search "<query>" --target T |
Start with esp32wb doctor . and paste its summary into the report; then do the
manual checklist below for what static tools cannot see.
Review checklist
Mark each item OK, ISSUE, or N/A, with file:line evidence for every ISSUE.
1. FreeRTOS safety
- Every task has an explicit stack size. In ESP-IDF the stack argument is in bytes (not words). Guide: ≥2048 simple, ≥4096 with Wi-Fi/BLE/printf-heavy code, ≥8192 for TLS — verify with high-water marks, not guesses.
- No blocking calls in ISRs; only
*FromISRAPIs;portYIELD_FROM_ISRused. - ISRs that run while flash cache is disabled are
IRAM_ATTRand touch only DRAM data. - Shared state protected by mutexes (priority inheritance), not by disabling interrupts.
- Spinlock critical sections (
portENTER_CRITICAL) are very short (no logging, no allocation). - No busy loops without blocking;
taskYIELD()notvTaskDelay(0). - Core affinity deliberate and valid for the target's core count (dual-core:
classic ESP32, S3, P4; S2 and the C/H series are single-core) — pin only to
cores that exist, or use
tskNO_AFFINITY. - Bounded timeouts on queues/semaphores where a hang would be a bug.
- Task watchdog: long operations feed it or are restructured.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today First seen · 125 lines · 124 tokens per session scan A 9d35271c55ac
esp32-arch-review is a skill published in the GitHub repository agodianel/esp32-claude-workbench (80 stars, last pushed yesterday), licensed MIT. It adds 124 tokens to every session and 1,579 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-29.
Other skills, from other repositories
connector-sdk-submit
Prepare a Community Connector or Fivetran Connector SDK example for a first pull request by applying the target repository's GitHub review instructions, sanitizing configuration.json, and running Black and Flake8.
kicad-pcb-review
Use this skill when reviewing, fixing, validating, or preparing KiCad PCB projects with the kicad MCP server.
kicad-pcb-review
KiCad PCB review and manufacturing readiness agent skill for Cursor.
rtl-review
Audit RTL code for lint violations, synthesis hazards, coding-style compliance, and readability. Use when the user says "review this Verilog", "check my RTL", "lint this module", "is this code synthesizable", or shares an HDL file and asks for feedback before simulation or tape-in.
vllm-rocm-pr-review
Review AMD/ROCm-related pull requests from the vllm-project/vllm GitHub repository and produce a single combined Chinese report: a detailed PR summary (summary, background & motivation, code-change analysis with Mermaid diagrams, technical principles, discussion highlights, risk table) followed by severity-sorted…
arch-review
Read-only architecture review of RTL vs uArch spec with area/timing/power tradeoffs. Use for post-RTL architecture sign-off or suspected spec mismatch.