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 agentlas-ai/agentlas-desktop --skill workspacegit clone --depth 1 https://github.com/agentlas-ai/agentlas-desktopWrote 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/agentlas-ai/agentlas-desktop/workspace)<a href="https://agentmods.dev/skills/agentlas-ai/agentlas-desktop/workspace"><img src="https://agentmods.dev/badge/skills/agentlas-ai/agentlas-desktop/workspace/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/agentlas-ai/agentlas-desktop/workspace"><img src="https://agentmods.dev/badge/skills/agentlas-ai/agentlas-desktop/workspace.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 | $0.00015 | $0.00692 |
| Opus 5.5 | $0.00006 | $0.00277 |
| Sonnet 5.5 | $0.00003 | $0.00138 |
| Haiku 4.5 | $0.00002 | $0.00069 |
Grade A, and why
workspace 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 2d 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 — 29 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Inside Agentlas Finance, use the corresponding short tool names from the current menu. Outside Finance, describe and call the namespaced finance.* harness operations below. The host validates the same account and source boundaries.
Dashboard and report workflow
Read finance.load, then describe finance.apply for the complete operation schema. Resolve IDs from the workspace, construct only the requested canonical operations, and pass its expectedRevision. Direct controls and these operations share the same persisted workspace. On a revision conflict reload and reconcile the intended change before retrying.
Widget period can inherit the dashboard period or remain pinned; preserve that distinction when changing a global period. Existing data and provenance come from provider retrieval, not authored operation payloads. New assets use finance.asset.add; source evidence uses finance.query.save.
finance.report.create freezes an existing dashboard using the host snapshot. finance.report.read inspects that saved snapshot. finance.apply can change report titles or section prose, reorder/remove sections, or explicitly refresh a report using its saved widget configuration and current stored data. Refresh creates a new report revision rather than silently changing evidence in the old report. Prose edits stay pending owner review. Explain any gap between a saved narrative and newer source observations.
To ground an existing report in another dataset, use finance.query.save with its active reportId, mode: "report" and current expectedRevision before revising the summary. This appends host-retrieved supporting sections while preserving existing sections, numerical projections and source bindings. Attach every dataset used in the prose to the report; a standalone source record does not freeze its data inside that report.
Set the requested BCP47 language when creating a report, or in a report.update patch when rewriting one. Write titles, headings and prose naturally in that language. Section presentation selects narrative, chart, an explicitly requested comparison, or supporting material. Keep original numerical projections intact. Favor a short conclusion and a relevant chart; a chart does not need its raw table repeated underneath. Use finance.calculate for validated allocation, historical scenarios and option calculations, with localized titles and interpretation in its optional presentation fields.
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.
- 2d ago Changed · +2 lines 00569010dc34
- 3d ago First seen · 27 lines · 15 tokens per session scan A 59f90110a26c
workspace is a skill published in the GitHub repository agentlas-ai/agentlas-desktop (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 15 tokens to every session and 692 once invoked, about $0.0001 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 2026-10-19.
Other skills, from other repositories
stripe-link-wallet
Agent wallet on the Stripe Link CLI (link-cli). Connect the wallet, then buy, purchase and pay for things on the user's behalf via approved spend requests, one-time-use cards, and 402 / Machine Payment Protocol (MPP) payments.
finance-balance-sheet
Generate a Balance Sheet (Statement of Financial Position) showing assets, liabilities, and owner equity at a point in time. Entity-aware equity treatment (sole-prop / partnership / LLC / S-corp / C-corp); GAAP vs tax-basis distinction. Templates only - not financial or tax advice.
finance-1099-prep
Vendor W-9 collection workflow, 1099-NEC vs 1099-MISC vs 1099-K decision tree, contractor classification check (IRS 20-factor + ABC test embedded), TIN matching, backup withholding triggers. Templates only - not personalized tax advice.
finance-payroll-prep
Quarterly Form 941 and annual Form 940 prep checklist, state UI / WC matrix, new-hire reporting, S-corp reasonable-salary documentation (Watson, Glass Blocks, Fleischer factors), and fringe-benefit valuation. Templates only - not personalized payroll, tax, or legal advice.
finance-cashflow
Generate a cash flow statement from net income and balance sheet changes. Routes accrual filers to indirect method, cash-basis filers to direct method. Covers operating, investing, and financing activities with period-end reconciliation.
finance-pl
Generate a Profit and Loss (P&L) statement from revenue and expense inputs, with period-over-period comparison and margin analysis. Basis-of-accounting-aware (cash / accrual / modified cash); ASC 606 deferred-revenue guidance for SaaS. Ported from Anthropic's financial-statements skill.