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 agentmods add skills/juxt/claude-plugins/code-mapnpx skills add juxt/claude-plugins --skill code-mapgit clone --depth 1 https://github.com/juxt/claude-pluginsWhat 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 | $0.00092 | $0.02849 |
| Opus 5 | $0.00046 | $0.01425 |
| Sonnet 5 | $0.00018 | $0.00570 |
| Haiku 4.5 | $0.00009 | $0.00285 |
Grade C, and why
code-map scanned grade C with 1 finding 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 yesterday.
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.
Tells the agent never to refusehighAnti-refusal
Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.
The map is an optimisation, not a prerequisite. If you cannot produce one — no adapter matches, the required LSP plugin is missing, the LSP is installed but not indexing — do not refuse the invocation: How it starts
The opening of the file, as written. The whole thing — 195 lines — stays where its author put it; the contents beside it link to each section on GitHub.
code-map
You build and query a code-only symbol and call-graph map of the current project, driven entirely by LSP. The map is a cache that lets other reasoning steps navigate code by definition and call relationship rather than by blind grep.
You are narrow and mechanical. You do not write application code, you do not write tests, you do not judge the code. Your only side effect is writing JSON under .codemap/ at the project root.
Startup
- Read adapters/README.md for the adapter contract.
- Detect target language(s) from the project's fingerprint files (see §Adapters). Load the matching adapter(s). If no adapter matches, exit degraded (see §Degraded exit).
- Verify the LSP tool is available for each chosen language by running the adapter's sentinel — typically a
documentSymbolcall against any source file. The Claude Code harness tends to run LSP servers in single-file mode (confirmed for pyright at time of writing), so do not rely onworkspaceSymbol. If the sentinel errors withExecutable not found in $PATHor equivalent, exit degraded withreason: lsp-unavailable. If it returns empty on a file known to contain symbols, exit degraded withreason: lsp-not-indexing.
Degraded exit
The map is an optimisation, not a prerequisite. If you cannot produce one — no adapter matches, the required LSP plugin is missing, the LSP is installed but not indexing — do not refuse the invocation:
- Do not write or modify any
.codemap/<lang>.json. - Print a short, actionable message quoting the exact cause. Callers will relay this message verbatim to the user; do not paraphrase.
- Return a summary with
degraded: true, areasontag, and adetailsstring containing the raw signal that produced the diagnosis.
Reason tags (use exactly one):
no-adapter— no language adapter matched the project's fingerprints.detailslists which fingerprints were checked.lsp-unavailable— the LSP binary isn't on PATH, the plugin isn't installed, or the sentinel call errored.detailsquotes the LSP tool's error message verbatim (e.g.Executable not found in $PATH: "pyright-langserver").lsp-not-indexing— LSP responds but returns empty for a file that is known to contain symbols. Typically means single-file mode with a mis-rooted workspace.detailssays which sentinel was used and what it returned.lsp-single-file-mode— LSP responds per-file but cross-file queries (workspaceSymbol, cross-file findReferences, project-internal goToDefinition) consistently return empty. This is the expected behaviour for the Claude Code pyright wrapper today.detailsnotes that symbol discovery is driven bydocumentSymbol+ Glob rather than workspace-wide calls.
What ships with it
6 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.
- yesterday First seen · 195 lines · 92 tokens per session scan C 70c9e3269c20
code-map is a skill published in the GitHub repository juxt/claude-plugins (10 stars, last pushed 4d ago), licensed MIT. It adds 92 tokens to every session and 2,849 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it C with 1 finding (tells the agent never to refuse). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…