map

A command for rebuilding an overview of several repositories and a map of the relationships between them.

In plain words
What is it for?
Use it to rewrite the portfolio and cross-repository map after reviewing repository knowledge cards, root dependency files, imports, URLs, shared paths, queues, and tables.
Why use it?
It keeps portfolio and cross-repository documentation aligned with the current code and distinguishes proven connections from merely similar projects.

Command for Claude Code

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 commands/end1989/repos-expert/map
Clone the repo
git clone --depth 1 https://github.com/end1989/repos-expert

Made for: Claude Code.

Per session 10 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 660 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 $0.00010 $0.00660
Opus 5 $0.00005 $0.00330
Sonnet 5 $0.00002 $0.00132
Haiku 4.5 $0.00001 $0.00066

Measured yesterday against content hash 6dad67622daf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

map 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 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.

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.

agent-kit/.claude/commands/map.md · 55 lines

What it actually says

Rewrite _knowledge/portfolio.md and _knowledge/cross-repo-map.md from the current card set.

Read every _knowledge/*/card.md, plus the root manifests of each repo (package.json, pyproject.toml, requirements.txt, go.mod, Cargo.toml, *.csproj) — those carry dependency facts the cards may have missed.

portfolio.md — what repos exist, one line each on what they are for, how they group into themes, and the overall state of the collection.

cross-repo-map.md — how the repos actually relate. The cards are summaries written by an earlier pass, not primary sources. You are sitting in the folder that holds every repository, with Read, Glob and Grep, so check anything that matters.

A claimed relationship is worth asserting only when you can point at what creates it: an import, a package dependency, a URL or host:port one side serves and the other calls, a shared file path, a queue or table name in both. Verify before asserting, and cite the file.

Keep three categories explicitly distinct, and never blur them:

  • Connected — evidenced in code, with the file cited.
  • Merely similar — same framework or conventions, no data or code flowing between them. Say so; convergent tooling is not a relationship.
  • Claimed but unconfirmed — a card or README implies a link you could not find. Say that you looked and what you found instead.

Two more findings are worth their own sections when they apply, because they are common here and neither is a "relationship":

  • Same codebase, two repo names — check with diff -rq <a>/src <b>/src before claiming it, and say whether the trees are identical or merely similar.
  • Referenced but absent — a repo depending on a sibling path that is not in this folder. That is a broken assumption worth surfacing, not an edge to draw.

"No evidence of a connection" is a useful, honest finding. An inferred link stated as fact is not.

With --diagram, also write _knowledge/wiring.mmd — a Mermaid flowchart of the connections. The grading must survive into the picture: solid edges for evidenced links, dashed for claimed-but-unconfirmed, and a distinct style for references to repos that are absent. Label each edge with its evidence (package.json:11, both bind :3001). Leave "merely similar" clusters out of the graph entirely and note them underneath — drawing them makes the map busier and mean less.

A diagram that renders all three grades as the same arrow is worse than no diagram: it is more persuasive than prose, and nobody re-reads a diagram skeptically.

No personal identifiers anywhere: no account names, emails, or remote URLs containing them.

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. yesterday First seen · 55 lines · 10 tokens per session scan A 6dad67622daf

Subscribe to this mod's changes

map is a command published in the GitHub repository end1989/repos-expert (0 stars, last pushed 2d ago), licensed MIT. It adds 10 tokens to every session and 660 once invoked, about $0.0001 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.