SurfSense is an open-source desktop application that uses an AI agent to research, transform, and edit documents while keeping their index on the user's machine. It is intended for people working with documents they cannot upload, with a choice of local or remote models and no account required.
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 MODSetter/SurfSense --skill pdfgit clone --depth 1 https://github.com/MODSetter/SurfSenseWrote 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/modsetter/surfsense/pdf)<a href="https://agentmods.dev/skills/modsetter/surfsense/pdf"><img src="https://agentmods.dev/badge/skills/modsetter/surfsense/pdf/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/modsetter/surfsense/pdf"><img src="https://agentmods.dev/badge/skills/modsetter/surfsense/pdf.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00029 | $0.00768 |
| Opus 5.5 | $0.00012 | $0.00307 |
| Sonnet 5.5 | $0.00006 | $0.00154 |
| Haiku 4.5 | $0.00003 | $0.00077 |
Grade A, and why
pdf 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 — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create the requested PDF in /workspace. Everything required is preinstalled;
never run pip install, npm install, or download dependencies.
Choose the renderer
- Use WeasyPrint for typography-heavy documents authored naturally as HTML/CSS: resumes, CVs, letters, one-pagers, and styled reports.
- Use ReportLab for programmatic layouts, drawing, charts, coordinates, or documents assembled directly from Python data.
- Use A4 unless the user or source context implies US Letter. Use margins of at least 18 mm (0.7 in), 10–11 pt body text, and 1.3–1.5 line height.
- Available families include DejaVu, Liberation, and Noto (including CJK). Prefer Liberation Sans/Serif for office-style documents and Noto for broad Unicode coverage.
- Do not embed page numbers, folios,
Page X of Ylabels, CSS page counters, or numbered ReportLab footers. SurfSense's PDF viewer supplies its own page indicator, so embedded numbering would duplicate the interface.
Escape untrusted text before placing it in HTML.
Converting a file the user already has
When the request is to turn an existing workspace file into a PDF, call
load_source_document(path="/documents/...") and convert the source_path it
returns. Re-authoring the content with a renderer above would discard the
original layout, so reach for it only when the source has no stored upload.
A source that is already a PDF needs no conversion. Convert Office formats with headless LibreOffice, giving each run its own profile directory so concurrent conversions cannot collide, and confirm the output exists before verifying:
soffice --headless -env:UserInstallation=file:///tmp/soffice-<unique> \
--convert-to pdf --outdir <out-dir> <source_path>
LibreOffice names the output after the source stem, so read it back from
<out-dir> rather than assuming your own filename.
Revisions
Start an in-place revision with load_artifact_for_revision. Read its
markdown_path for the artifact's substantive content and any other context
provided by the user, then regenerate the PDF at expected_output_path. PDF
revisions are rebuilds from Markdown/context, not edits inferred from rendered
pages. Do not use vision to reconstruct the old PDF; vision may still be used
by verify_artifact after regeneration.
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 Changed a17728addfb1
- 1mo ago First seen · 75 lines · 29 tokens per session scan A 567e8aac413e
pdf is a skill published in the GitHub repository MODSetter/SurfSense (16,464 stars, last pushed today), licensed Apache-2.0. It adds 29 tokens to every session and 768 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 2026-08-30.
Other skills, from other repositories
leiloeiro-edital-ranbot-ai
Analise e auditoria de editais de leilao judicial e extrajudicial. Riscos ocultos, clausulas perigosas, debitos, ocupante e classificacao da oportunidade.
first-reader
Beta readers for any draft, run by simulating how a real reader experiences it, moment by moment. A skim gate, a no-lookahead timed read producing an attention transcript with quit points, a recall test of what a reader remembers the next day, and a trust ledger of the implied author. Use when the user asks for a beta…
mcp-apps-builder
MANDATORY for ALL MCP server work - mcp-use framework best practices and patterns. READ THIS FIRST before any MCP server work, including: Creating new MCP servers Modifying existing MCP servers (adding/updating tools, resources, prompts, widgets) Debugging MCP server issues or errors Reviewing MCP server code for…
project-graveyard
Scans the developer's machine for dead side projects, autopsies each one from its git history (died at the payments wall, killed by a newer project, finished but never shipped), surfaces their personal death patterns, and picks the corpse most worth resurrecting — then helps ship it. Use when the user mentions…
commit-archaeologist
Reconstructs why code exists from local git history, including the introducing commit, later changes, current authors, repeated companion files, and likely intent. Use when the user asks "why does this code exist", "who wrote this function and why", or to "explain the history of this function" before a rewrite…
dependency-doctor
Checks requirements.txt, pyproject.toml, and package.json dependency manifests for surface-level direct-dependency footguns: standard-library shadowing pins, abandoned backports, unpinned dependencies, and obvious intra-manifest conflicts, plus opt-in PyPI yanked releases. Use when the user asks to check a manifest…