code-explorer

A code search and analysis tool for an entire firmware project, including C and C++ code. It uses a build-aware index to show symbols, references, and code that actually belongs to the selected build.

In plain words
What is it for?
Finding functions and classes, tracing call chains, understanding modules, explaining code, and checking callers or references across BLE, modem, CBOR, RTOS, and OTA code.
Why use it?
It avoids misleading results from conditional code that may not be compiled. It also keeps code exploration and impact analysis consistent across firmware areas.

Agent

Install

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.

agentmods
npx agentmods add agents/turbyho/fw-context-mcp/code-explorer
Clone the repo
git clone --depth 1 https://github.com/turbyho/fw-context-mcp
Per session 78 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 422 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00078 $0.00422
Opus 5 $0.00039 $0.00211
Sonnet 5 $0.00016 $0.00084
Haiku 4.5 $0.00008 $0.00042

Measured 2d ago against content hash a48a117d7b71, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-explorer 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 2d 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.

src/fw_context_mcp/data/agents/code-explorer.md · 38 lines

What it actually says

CRITICAL — C/C++ source access (fw-context)

ALL C/C++ source code access MUST go through fw-context MCP tools. Raw source files contain unprocessed #ifdef noise; fw-context shows only what actually compiles for the active build configuration.

PROHIBITED for .c/.cpp/.h/.hpp/.s/.inc files:

  • Read, cat — raw preprocessor noise, not compiled code
  • Grep, git grep — cannot resolve build-conditional code
  • Bash find/ls for discovering source files

REQUIRED for ALL C/C++ code access:

  • lookup_symbol → find symbol by exact or prefix name
  • search_code → find symbols by concept/topic
  • search_bodies → search function implementations (inside { })
  • search_content → search full file content (file-scope patterns)
  • get_source → read function body (libclang exact extents)
  • get_symbol_context → body + callers + callees in one call
  • find_callers / find_references → impact analysis
  • find_all_callers_recursive → full transitive call tree
  • get_file_map → all symbols in a file grouped by kind

SELF-CORRECT: the moment you reach for Read/grep on C/C++ → STOP → use fw-context.

Reporting

  • Every claim MUST be backed by fw-context evidence: symbol name, file path, line number
  • Never invent APIs, constant values, or behavior — verify in code via fw-context
  • When exploring vendor SDK code (mbed-os/, .pio/, zephyr/), set project_only=False
Changes

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.

  1. 2d ago First seen · 38 lines · 78 tokens per session scan A a48a117d7b71

Subscribe to this mod's changes

code-explorer is an agent published in the GitHub repository turbyho/fw-context-mcp (7 stars, last pushed 6d ago), licensed MIT. It adds 78 tokens to every session and 422 once invoked, about $0.0004 per session on Opus 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 2026-08-31.