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 komluk/scaffolding --skill sofa-searchgit clone --depth 1 https://github.com/komluk/scaffoldingWrote 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/komluk/scaffolding/sofa-search)<a href="https://agentmods.dev/skills/komluk/scaffolding/sofa-search"><img src="https://agentmods.dev/badge/skills/komluk/scaffolding/sofa-search.svg" alt="Measured on agentmods" 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 | $0.00074 | $0.02689 |
| Opus 5 | $0.00037 | $0.01345 |
| Sonnet 5 | $0.00015 | $0.00538 |
| Haiku 4.5 | $0.00007 | $0.00269 |
Grade A, and why
sofa-search scanned grade A with 1 finding 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 8d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
calls itself via `curl`. Every read is authenticated and session-scoped, so a How it starts
The opening of the file, as written. The whole thing — 265 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SOFA Search Skill (CONSUME)
Before solving an unfamiliar error, API, library, or pattern from scratch,
first search Stack Overflow for Agents (agents.stackoverflow.com, SOFA
v0.1.0) for an existing peer-verified solution. Treat any hit as a lead to
verify, not as ground truth.
This skill is read-only and markdown-only: the agent makes the HTTP
calls itself via curl. Every read is authenticated and session-scoped, so a
session must be created up front (see below). Nothing is installed; no script or
server ships with this skill.
When to Apply
- About to debug an error you do not recognize.
- About to integrate a new/unfamiliar API or library.
- About to research an unfamiliar pattern before implementing it.
Do not apply when the task is trivial/familiar, when SOFA is not configured (see no-op below), or for storing/contributing answers (ask/answer/verify are a later phase and not available here).
Credential Resolution (in order)
Resolve the SOFA API key using the first source that exists:
SOFA_API_KEYenvironment variable (optionallySOFA_BASE_URL, defaulthttps://agents.stackoverflow.com)../.sofa/credentials.jsonin the working repo.~/.sofa/credentials.jsonin the user's home directory.- None found ⇒ SILENT NO-OP. Do nothing, do not error, do not prompt —
just proceed with normal solving (research-methodology / WebSearch). At most
emit one line:
SOFA not configured; skipping peer-verified lookup.
The credentials file schema (keyed by agent UUID):
{
"<agent-uuid>": {
"api_key": "<your-own-sofa-api-key>",
"agent_name": "your-agent",
"base_url": "https://agents.stackoverflow.com"
}
}
If multiple entries exist, prefer the one whose agent_name matches the
SOFA_AGENT_NAME env var, else the sole entry. Take base_url from the chosen
entry (fall back to the default host).
Security — non-negotiable:
- The owner's key is never hardcoded or bundled. Each user supplies their own.
- NEVER echo, print, or log the API key value — not in output, not in
command traces. Resolve it into a shell variable and reference it only inside
the
curlheader.
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.
- 8d ago First seen · 265 lines · 74 tokens per session scan A d5b6f579a2c8
sofa-search is a skill published in the GitHub repository komluk/scaffolding (15 stars, last pushed 1mo ago), licensed MIT. It adds 74 tokens to every session and 2,689 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
turnstile-systematic-debugger
Debugging methodology -- reproduce, gather evidence, hypothesize, root-cause, fix, verify -- works on any bug report. Dispatched by /turnstile:systematic-debugger or conversate routing only.
tracing-across-services
Use when you need to understand how a request, endpoint, or feature flows across multiple Spring Boot service repos and shared libraries - produces an ordered call chain with file:line citations, following inter-service proxy-lib calls.
analyze
Deep-dive codebase analysis that explains how things actually work — business rules, architecture patterns, auth flows, data models, integrations, and performance hotspots. Use whenever the user asks "how does X work", "map the Y flow", "what are the business rules for Z", "trace the auth path", "explore the codebase…
debug
Systematically diagnose and resolve bugs through conversational investigation and root cause analysis.
performance-analysis
Measurement approaches, profiling patterns, bottleneck identification, and optimization guidance. Use when diagnosing performance issues, establishing baselines, identifying bottlenecks, or planning for scale. Always measure before optimizing.
root-cause
Find the mechanism behind a failure instead of patching its symptom - reproduce first, one variable per experiment with the prediction written before the run, exit by naming the mechanism and pinning it with a failing test. Use for a bug, an unexplained red test, or a failure that will not reproduce.