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 Michael-Obele/docshark --skill using-docsharkgit clone --depth 1 https://github.com/Michael-Obele/docsharkWrote 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/michael-obele/docshark/using-docshark)<a href="https://agentmods.dev/skills/michael-obele/docshark/using-docshark"><img src="https://agentmods.dev/badge/skills/michael-obele/docshark/using-docshark/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/michael-obele/docshark/using-docshark"><img src="https://agentmods.dev/badge/skills/michael-obele/docshark/using-docshark.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 | $0.00035 | $0.01174 |
| Opus 5.5 | $0.00014 | $0.00470 |
| Sonnet 5.5 | $0.00007 | $0.00235 |
| Haiku 4.5 | $0.00003 | $0.00117 |
Grade A, and why
using-docshark 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 — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Operating DocShark
Start Modes
STDIO mode (MCP clients)
docshark --stdio
or
npx docshark --stdio
Use STDIO when a client launches DocShark as a subprocess MCP server.
HTTP mode
docshark
Default HTTP port is 6380 (override with --port).
HTTP endpoints:
/mcp/sse/api/api/health
Data Directory
- Default data directory:
~/.docshark - Override with:
- CLI:
--data-dir <path> - Env:
DOCSHARK_DATA_DIR
- CLI:
Core CLI Operations
docshark add <url>docshark listdocshark search <query>docshark refresh <name>docshark stale [--days <n>](alias:outdated) — list libraries older than the freshness windowdocshark icons [style](alias:icon) — show or set the CLI icon styledocshark rename <current-name> <new-name>docshark remove <name>docshark get [url] --library <name> --path <path>
Stale-Docs Auto-Prompt
- A library is stale after 14 days without a crawl. Tune with
DOCSHARK_STALE_DAYS=<n>,docshark stale --days <n>, or the--daysflag (clamped to 1–365). docshark listanddocshark staleon an interactive TTY print every stale library with its last crawl date and age, then askRefresh N stale libraries now? [y/N]. Answeringyre-crawls all listed libraries sequentially.- Non-interactive runs (pipes, CI) never prompt: they print the stale list on stderr with a hint instead.
docshark searchprints a one-line stale reminder (interactive only).- Skips:
docshark list --no-stale-check, orDOCSHARK_DISABLE_STALE_CHECK=1. - AI parity: the same signal is exposed over MCP via the Age/⚠️ columns of
list_librariesandmanage_library action=stale, so an assistant can tell the user which docs are outdated and refresh them after agreement.
Icon Styles
docshark icons [style]shows or sets the icon style; sets persist to<data dir>/config.json. Valid styles:emoji | nerd | plain | none.- Precedence:
--icons <style>/-Iflag (per run) →DOCSHARK_ICONSenv → config file → defaultplain. - Default
plain= common-monospace Unicode (✓ ✗ ⚠ ↻ …); any icon a style lacks (e.g. the shark) falls back to the emoji automatically. nerd= crisp monochrome glyphs from Nerd Fonts (Material Design Icons set); requires a Nerd Font-patched terminal font — otherwise glyphs show as boxes.none= text only (good for logs). Invalid values warn once and fall through to the next source.- MCP tool output is always emoji, since AI chat clients render those reliably — these settings only affect terminal output.
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 · +49 lines 0f34c3dd1f1f
- 1mo ago First seen · 69 lines · 35 tokens per session scan A 473629cd9fd3
using-docshark is a skill published in the GitHub repository Michael-Obele/docshark (2 stars, last pushed 1mo ago), licensed MIT. It adds 35 tokens to every session and 1,174 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 2026-08-31.
Other skills, from other repositories
init-workspace-documentation
Skill "init-workspace-documentation" from griddynamics/rosetta, covering agent memory.md, agent memory, preventive rules, what worked and what failed.
servicenow-sdk-docs-builder
Automation skill to generate or update the servicenow-sdk-docs documentation skill by fetching and processing the ServiceNow SDK and Examples repositories. It combines READMEs, extracts code samples, harvests the SDK CLI/auth docs, and creates a self-contained, portable documentation skill.
alcove
HTTP API-based documentation server (always running). Questions about project architecture, conventions, decisions, code structure, tech debt, env config, progress, or doc health. Also: store and recall agent memory (global memory/ + per-project /memory/ inside the docs vault), init project, audit docs, lint…
init
Bootstrap docaudit for a repo, or add and refresh its optional local documentation harness. Use when the user asks to set up docaudit, initialize the doc-audit harness, or run docaudit on a repo that lacks /check-docs / /review-docs / doc-lint. Inventories the repo, proposes a doc-audit.json config for approval…
zig-docs-mcp
Connect Prime Agent to the local zig-docs-mcp MCP server for always-fresh official Zig documentation of the latest release, std-library symbol docs from released sources, curated high-performance lightweight-software guidance, and safe local Zig toolchain auto-update prompts. Use when writing, reviewing, or debugging…
ovellum-docs
Set up and maintain Ovellum documentation for a TypeScript or JavaScript project — scaffold, build, validate, and safely edit generated docs whose hand-written prose survives regeneration. Use when the user wants to create, build, or maintain docs with Ovellum, or asks about protected zones, orphans, llms.txt, or the…