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 web3dev1337/agent-workspace --skill repo-atlasgit clone --depth 1 https://github.com/web3dev1337/agent-workspaceWrote 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/web3dev1337/agent-workspace/repo-atlas)<a href="https://agentmods.dev/skills/web3dev1337/agent-workspace/repo-atlas"><img src="https://agentmods.dev/badge/skills/web3dev1337/agent-workspace/repo-atlas.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00097 | $0.02360 |
| Opus 5 | $0.00048 | $0.01180 |
| Sonnet 5 | $0.00019 | $0.00472 |
| Haiku 4.5 | $0.00010 | $0.00236 |
Grade A, and why
repo-atlas 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 4d 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 — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repo Atlas
One queryable map of every repo the user owns — including repos that are not cloned on this machine. Ask it before searching the filesystem.
Why this exists
There are hundreds of repos. Grepping ~/GitHub finds only the fraction that happen to be cloned, costs thousands of tokens, and cannot tell you that a scruffy prototype has the best test harness in the collection. The atlas answers both "where is it?" and "is it worth copying?".
The one command that matters
atlas find <topic> # who did this well, and where in the repo
Example:
$ atlas find data-compression
5/5 acme-tycoon
bitpacked player save — 12x smaller than the JSON we started with
paths: src/data/packSave.ts
~/GitHub/acme-tycoon
Results are ranked by quality (1–5, recorded per topic), and repos the user has explicitly marked do-not-copy for that topic are excluded. ⚠old means untouched for over a year — still readable, just check it against current conventions.
Reading the map
atlas digest # compact whole-map overview — cheap, paste-able
atlas topics # every topic anyone has recorded, and who has it
atlas show <id> # everything known about one repo
atlas evidence <id> # live history, code/test counts, and representative paths
atlas list --platform steam --no-forks
atlas find testing --min-quality 4
Use atlas evidence <id> --json before describing why a local repo matters. When the entry has a GitHub identity, the command verifies the checkout's origin before inspection. It reports measured facts and keeps source contents and absolute machine paths out of the result. A frequently-changed-source path is a place to inspect, not a quality rating. Curated highlights remain the only quality judgement.
atlas digest is the right first call when you want orientation rather than an answer. It is deliberately terse:
steam physics-kit(physics:5, testing:5) puzzle-proto(testing:4 ⚠old)
example acme-tycoon(data-compression:5, worldgen:4)
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.
- 4d ago Changed · +56 lines eba7cd25532a
- 8d ago First seen · 119 lines · 97 tokens per session scan A 7f6f37f95ac9
repo-atlas is a skill published in the GitHub repository web3dev1337/agent-workspace (35 stars, last pushed 4d ago), licensed MIT. It adds 97 tokens to every session and 2,360 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
automation-flows
Use when building or fixing a no-code automation on n8n, Make, or Zapier — trigger to multi-app steps with data mapping, dedup, retries and an error path — or picking the platform by billing unit (task vs credit vs execution). NOT a typed API client in code (that is api-connector-builder), NOT a webhook receiver in…
code-dispatch
Delegate coding, file-system, and research tasks to a real CLI agent running on the user's machine. The orchestrator stays clean — no token overhead, no context window bloat. The specialist executes.
context-optimizer
Analyze skill content for optimal placement (Skill vs Passive Context vs Hybrid), compress markdown to pipe-delimited format (60-80% token reduction), and validate compliance against the decision framework. Passive context wins for what the model cannot know (post-cutoff APIs, repo gotchas); pre-trained knowledge…
handoff
Checkpoint work for recovery — fires on "handoff", "wrap up", context pressure, a closed phase, or before a long external wait or irreversible gate. Part of the Agentsmith harness; writes durable state plus a paste-ready kickoff block so a fresh session loses nothing.
analysis-provenance
Identify code ownership before modifying validators or linters. Checks file headers for provenance indicators, reviews documentation, and determines provenance as UPSTREAM, LOCAL, VENDOR, or UNKNOWN. Prevents accidental modification of upstream tools.
panning-for-gold
Triage raw unstructured input (transcripts, brain dumps) into evaluated thread inventories and a synthesized gold-found file. Four phases: front-gate, extract, evaluate, synthesize. Use when you say "pan for gold", "triage transcript", "synthesize gold-found", or hand it a voice transcript or brain dump. Do NOT use…