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 turbyho/fw-context-mcp --skill fw-reviewgit clone --depth 1 https://github.com/turbyho/fw-context-mcpWrote 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/turbyho/fw-context-mcp/fw-review)<a href="https://agentmods.dev/skills/turbyho/fw-context-mcp/fw-review"><img src="https://agentmods.dev/badge/skills/turbyho/fw-context-mcp/fw-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/turbyho/fw-context-mcp/fw-review"><img src="https://agentmods.dev/badge/skills/turbyho/fw-context-mcp/fw-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 | $0.00000 | $0.03489 |
| Opus 5 | $0.00000 | $0.01744 |
| Sonnet 5 | $0.00000 | $0.00698 |
| Haiku 4.5 | $0.00000 | $0.00349 |
Grade A, and why
fw-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 6d ago.
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 — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scope
- Review of diffs, commits, or specific changed files in embedded C/C++ projects
- Bugs, behavioral regressions, runtime risks, concurrency, memory, and platform rule violations
- Changes outside firmware code (docs, CI, scripts, plans) — skip fw-context, review manually
CRITICAL — MANDATORY
For C/C++ source code, use ONLY fw-context tools for reading, searching, and browsing files.
MUST NOT use any other tool to read or search C/C++ source. fw-context provides
ifdef-filtered code — only what actually compiles for the active build configuration.
Other tools show unprocessed preprocessor noise including dead #ifdef branches.
MUST NOT skip Phase 0. Every review MUST begin by creating and writing out the plan.
Core principle — verify in full project context
Every code change MUST be verified recursively against the entire project, not just the local diff fragment. A change that looks correct in isolation can be wrong in the broader context. For every changed symbol (function, type, macro):
- Callers — who calls this? Were they all updated? Use
find_callersthenfind_all_callers_recursivefor the full call tree. - Callees — what does this call? Is the new behavior still compatible with downstream
dependencies? Use
get_symbol_contextorfind_callees_recursive. - Users — who reads/writes this type or variable? Use
find_references. - Duplicates — does a similar implementation exist elsewhere? Use
search_bodiesand scan ALL results. - Removals — is the removed symbol truly gone from every file? Use
search_contentto confirm zero results project-wide. - Dead code — did the removal leave orphaned functions? Use
find_dead_code.
The diff shows SCOPE. fw-context verifies CORRECTNESS. Never conclude a review from the diff alone.
Phase 0: Create a review plan (MANDATORY)
BEFORE any detailed analysis, classify EVERY changed file and build a checklist of fw-context tools to run against each changed symbol. This is the single most effective guard against missed checks.
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.
- 6d ago Changed 3b42f4398a1c
- 11d ago First seen · 231 lines · 0 tokens per session scan A c9f742b54798
fw-review is a skill published in the GitHub repository turbyho/fw-context-mcp (8 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,489 tokens. 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 2026-08-31.
Other skills, from other repositories
kicad-pcb-review
Use this skill when reviewing, fixing, validating, or preparing KiCad PCB projects with the kicad MCP server.
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.
cdc-tool-profiles
Internal reference: cdc tool profiles (agent-loaded; do not invoke).
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.
kicad-pcb-review
KiCad PCB review and manufacturing readiness agent skill for Cursor.
code-review-policy
Internal reference: code review policy (agent-loaded; do not invoke).