Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/bgmacris/quimera-ainpx agentmods add skills/bgmacris/quimera-ai/mapWrote 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/bgmacris/quimera-ai/map)<a href="https://agentmods.dev/skills/bgmacris/quimera-ai/map"><img src="https://agentmods.dev/badge/skills/bgmacris/quimera-ai/map/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/bgmacris/quimera-ai/map"><img src="https://agentmods.dev/badge/skills/bgmacris/quimera-ai/map.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 | $0.00098 | $0.02976 |
| Opus 5.5 | $0.00039 | $0.01190 |
| Sonnet 5.5 | $0.00020 | $0.00595 |
| Haiku 4.5 | $0.00010 | $0.00298 |
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 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 — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.
tandem:map — per-site navigation memory
Persists how a specific site is navigated in a per-host profile. Avoids re-deriving the
structure (huge snapshots, ephemeral refs) every session. Complements the
"Real navigation — patterns" section of the tandem skill: that is CROSS-site technique (infinite
scroll, cookie banners…); this is knowledge OF-a-site (the routes of THAT site).
Full design and rationale: tandem/docs/01-navigation-memory.md.
Store (engine: scripts/map.sh)
Profiles in ~/.claude/tandem/sites/<host>.md (OUTSIDE git; override TANDEM_DATA_DIR).
The SOURCE OF TRUTH are the .md files; the index is DERIVED from their frontmatter, not kept separately.
map.sh list............ table of profiles (host · updated · file)map.sh show <host>..... dumps the profile (accepts a full URL: normalizes to host)map.sh path <host>..... the file path (whether it exists or not) — for reading/writingmap.sh index........... derived JSON index (to stdout + cachessites/index.json)
Invoke it by its path within the plugin (scripts/map.sh); there's no bin on PATH for map.
When to act (reflexes)
- When you start operating a site: if a profile already exists, it usually reaches you on its
own — a
PostToolUsehook onbrowser_navigateinjects the host's profile the first time you navigate to it in the session (scripts/hook-inject-profile.mjs). If it didn't arrive (or you want to re-read it):map.sh show <host>. If there's no profile and the work isn't trivial → offer the human to do a recon and save it. - Before trusting a route's locators (cheap re-check): validate with
fingerprint.mjs check <host> <route>(structural signal, small JSON), NOT with a targetedbrowser_snapshot(it weighs ~18× the whole profile).match→ trust that route'ssel:values WITHOUT a snapshot (see "Frugal flow").drift→ point 4. No fingerprint yet → a one-off targeted snapshot, and capture the fingerprint for next time. - After a recon worth keeping: draft the profile, show it to the human, and only with their OK write it.
- If a locator fails (drift): do NOT try to repair it. Re-derive live, update the profile line with the new anchor and today's date, and tell the human about the change. Rule: a wrong cached click is worse than a slow click.
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 · 180 lines · 98 tokens per session scan A cceda17a7775
map is a skill published in the GitHub repository bgmacris/quimera-ai (12 stars, last pushed 3mo ago), licensed MIT. It adds 98 tokens to every session and 2,976 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
chrome-agent
Local browser automation with structured, verified outcomes. Use for web navigation, scraping and extraction, form interaction, screenshots and downloads, network or console checks, responsive testing, or page-scoped device emulation.
scrape-structured-data
Get the repeating records off a web page (product grids, search results, job listings, news feeds, tables) as JSON, without writing CSS selectors and without spending a model call to read the HTML. Works on sites with no API, including ones behind a login or bot protection. Runs locally, one binary, no API key. Use…
cdpilot
Browser automation over raw Chrome DevTools Protocol — navigate, click, fill forms, screenshot, extract content, handle iframes, and bypass basic bot detection. Use whenever a task needs a real browser (no Puppeteer/Playwright/Selenium available, or when those get detected) — for example "open this page and click X"…
owb
Open Web Bridge (OWB) — drive the user's own real browser with the owb command. Read pages behind their existing logins, gather and cross-check information, fill forms, walk multi-step flows, debug their site, audit responsive/accessibility behavior, and capture or reverse-engineer network traffic. Use this whenever…
browser
Use this skill when the user says browser, /browser, test in Chrome, inspect a webpage, verify a localhost app, capture screenshots, check console/network errors, run browser QA, or automate browser flows with the Mochi browser MCP.
verify-chrome-cdp-ex
Verify chrome-cdp-ex (CLI and MCP driving a real Chrome). Use when proving a fix, or when a reviewer re-runs the user path on a PR head.