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 Prajwalsrinvas/claude-code-reverse-engineering --skill skillgit clone --depth 1 https://github.com/Prajwalsrinvas/claude-code-reverse-engineeringWrote 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/prajwalsrinvas/claude-code-reverse-engineering/skill)<a href="https://agentmods.dev/skills/prajwalsrinvas/claude-code-reverse-engineering/skill"><img src="https://agentmods.dev/badge/skills/prajwalsrinvas/claude-code-reverse-engineering/skill/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/prajwalsrinvas/claude-code-reverse-engineering/skill"><img src="https://agentmods.dev/badge/skills/prajwalsrinvas/claude-code-reverse-engineering/skill.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 | $0.00044 | $0.02726 |
| Opus 5.5 | $0.00018 | $0.01090 |
| Sonnet 5.5 | $0.00009 | $0.00545 |
| Haiku 4.5 | $0.00004 | $0.00273 |
Grade A, and why
analyze-cc-feature 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 1mo 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 — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Analyze Claude Code Feature
Reverse engineer a specific feature in Claude Code's source code.
Terminology note: Claude Code's cli.js is minified (whitespace removed, identifiers shortened for size), NOT obfuscated (intentionally made hard to read with control flow flattening, string encryption, etc.). webcrack's --no-deobfuscate flag is used because there is no obfuscation to reverse. The output file is called deobfuscated.js by webcrack convention, but it is more accurately "unminified" source.
IMPORTANT: Use the task list (TaskCreate/TaskUpdate) to track progress through the pipeline. Create all tasks upfront, then work through them sequentially, marking each in_progress when starting and completed when done.
Pipeline
Create these tasks at the start:
- Acquire source — Extract cli.js from the installed native Bun binary
- Unminify — Run webcrack + prettier to produce readable source
- Locate the feature — Search for anchor strings to find the code
- Extract & annotate — Pull out the feature's code, rename identifiers inline
- Analyze — Document what the feature does step by step
- Write report — Produce a markdown report with evidence
- Self-reflect — Evaluate whether the skill itself should be updated
If the user says to skip steps (e.g., "unminified source already exists at X, skip steps 1-2"), create those tasks as completed immediately.
Step 1: Acquire source
Run scripts/extract-cli.sh to extract cli.js from the installed native binary. The script caches by version.
bash ~/.claude/skills/analyze-cc-feature/scripts/extract-cli.sh
Important — the source is no longer in the npm package. Since v2.1.113 the npm package (@anthropic-ai/claude-code) is a ~170 KB thin installer; the real CLI ships as a per-platform native Bun standalone executable (~250 MB) with the JavaScript embedded in its module graph. extract-cli.sh locates the installed binary (~/.local/share/claude/versions/<version> or $(which claude)) and runs scripts/extract-bun-cli.py to pull cli.js out of the binary's module graph. It also drops the sibling modules (a precompiled Bun bytecode blob and two Rust NAPI addons — image-processor.node, audio-capture.node) into /tmp/claude-code-npm/bun-modules/.
What ships with it
5 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.
- 1mo ago First seen · 167 lines · 44 tokens per session scan A 0c4326fa27de
analyze-cc-feature is a skill published in the GitHub repository Prajwalsrinvas/claude-code-reverse-engineering (3 stars, last pushed 3mo ago), licensed MIT. It adds 44 tokens to every session and 2,726 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
generative-thinking
Break out of a locked problem frame by picking one disciplined move — reframe, provocation (Po), random stimulus, SCAMPER, inversion, structured analogy, constraint play, or family traversal — and committing to it before evaluating. Use when stuck, when options feel narrow or obvious, when iterations produce…
swarming-to-green
Runs a long swarm of hundreds to thousands of cheap headless coding workers that drive a large failing-test backlog to green. Failures are clustered by signature, each worker fixes one cluster in its own git worktree, a serial no-regression merge gate lands the fixes, the loop resumes across restarts for hours, and it…
declauding
Load at the start of any task whose deliverable is prose another person will read — a PR description, commit message, README, doc, postmortem, blog post, issue or review comment, report, release note or essay — before drafting it, whether the result is pushed, posted, published or saved to a file, without being asked.…
developing-preact
Specialized Preact development skill for standards-based web applications with native-first architecture and minimal dependency footprint. Use when building Preact projects, particularly those involving data visualization, interactive applications, single-page apps with HTM syntax, Web Components integration, CSV/JSON…
orchestrating-chat-agents
Fans Claude work out over the Anthropic API from surfaces with no native subagents, chiefly claude.ai chat and projects, with parallel calls with a shared cached context, streaming, long-lived conversation threads with cached history, and named agent pools; also delegates to Gemini through the Cloudflare AI Gateway on…
browsing-bluesky
Browse Bluesky content via API and firehose - search posts, fetch user activity, sample trending topics, read feeds and lists, analyze and categorize accounts. Supports authenticated access for personalized feeds. Use for Bluesky research, user monitoring, trend analysis, feed reading, firehose sampling, account…