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 agentmods add skills/mcwalrus/oppertunity-party/tech-guidenpx skills add mcwalrus/oppertunity-party --skill tech-guidegit clone --depth 1 https://github.com/mcwalrus/oppertunity-partyWhat 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 | $0.00133 | $0.01850 |
| Opus 5 | $0.00067 | $0.00925 |
| Sonnet 5 | $0.00027 | $0.00370 |
| Haiku 4.5 | $0.00013 | $0.00185 |
Grade A, and why
tech-guide 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 yesterday.
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 — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Technical Guide
Explore a repository or technology through structured, live filesystem discovery and cited research. The agent plans the exploration path, maps the landscape with tree, ls, and cat, and when knowledge gaps appear, searches the live web for current documentation rather than presenting pre-training memory as fact.
Values
- Live over remembered. Never describe filesystem structure or file
contents from memory. Run
tree,ls -al, orcateach time. - Cited over claimed. Every factual statement about a technology, dependency, or version must carry a citation or a live lookup. If you cannot fetch it, say so explicitly — do not fabricate.
- Planned over ad-hoc. Open every tour with
creative-plannerto build an exploration plan. Adapt the plan as findings change, but never wander without a map.
Constraints
- Never present pre-training-era recall as fact about a technology's current behaviour, version, or status.
- Never fabricate file contents. If a file cannot be read, say so and move on.
- Never output a long list of files unprompted. Every observation must connect the current file to the broader system.
- When exploring a local repository, probe the filesystem before searching the web. First look at what exists; then research what you do not understand. When the user names a technology without a local repository, web research comes first.
Opening
Establish what the user wants to explore and their role before reading files:
"What are you trying to understand — a new codebase you've joined, a technology you're adopting, or a specific subsystem? What's your role here (engineer, contributor, reviewer), and what do you most need to know?"
If they name a technology without a local repository (e.g. "guide me
through Docker"), do not attempt filesystem probing. Instead, use WebSearch
to find the technology's latest official documentation, then read it with
WebFetch. Build the tour from cited sources, not training memory.
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.
- yesterday First seen · 177 lines · 133 tokens per session scan A 6a5308dbd7f4
tech-guide is a skill published in the GitHub repository mcwalrus/oppertunity-party (0 stars, last pushed 9d ago), licensed MIT. It adds 133 tokens to every session and 1,850 once invoked, about $0.0007 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 2026-08-31.
Other skills, from other repositories
agent-browser
Browser automation for testing and verification. Use when you need to interact with web UIs, verify visual changes, fill forms, or capture screenshots.
repro-admin
Reproduce an EmDash admin UI bug. Attach a container, start the demo dev server, drive the admin with agent-browser using the dev-bypass session, and capture the reproduction as screenshots plus a replayable transcript.
repro-public
Reproduce a bug in the public-facing rendered site (not the admin). Attach a container, start the demo dev server, drive public routes with agent-browser, and capture the reproduction as screenshots plus a replayable transcript.
wp-static-clone
Clones a live WordPress (or other CMS-driven) site into a static HTML site deployable on any static host (Cloudflare Pages, Netlify, Vercel, S3+CloudFront, plain Apache/nginx). Use when the user wants to "scrape", "freeze", "archive", "static-ify", or "move to [host]" a WordPress site, or asks to turn a sitemap into…
astro-best-practices
This skill should be used when working on any Astro project or when the user asks about Astro best practices, conventions, or patterns. Trigger when: working on an Astro project, creating Astro components/pages/layouts, configuring astro.config, setting up Content Collections, using Content Layer API, choosing client…
chrome-ext-best-practices
This skill should be used when working on any Chrome extension project or when the user asks about Chrome extension best practices, conventions, or patterns. Trigger when: setting up a new Chrome extension, configuring WXT, asking about MV3 architecture, "Chrome extension project structure", "WXT config", "extension…