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 wasintoh/toh-framework --skill engineer-harnessgit clone --depth 1 https://github.com/wasintoh/toh-frameworkWrote 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/wasintoh/toh-framework/engineer-harness)<a href="https://agentmods.dev/skills/wasintoh/toh-framework/engineer-harness"><img src="https://agentmods.dev/badge/skills/wasintoh/toh-framework/engineer-harness/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/wasintoh/toh-framework/engineer-harness"><img src="https://agentmods.dev/badge/skills/wasintoh/toh-framework/engineer-harness.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Privilege Escalation · line 122 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
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 | $0.00109 | $0.02736 |
| Opus 5.5 | $0.00044 | $0.01094 |
| Sonnet 5.5 | $0.00022 | $0.00547 |
| Haiku 4.5 | $0.00011 | $0.00274 |
Grade A, and why
engineer-harness 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 — 234 lines — stays where its author put it; the contents beside it link to each section on GitHub.
🛠️ Engineer Harness Skill
Purpose: Shared harness for the main commands — pick tools like a senior engineer, talk like a human, suggest what's next Version: 1.1.0 For: Toh Framework v2.0.0+ Used by:
/toh,/toh-plan,/toh-fix,/toh-vibe(main commands) — MANDATORY · pairs withorchestration-protocolReplaces: the two legacy reporting skills (human report + next-step suggestions), now merged
🎯 Purpose
Three things every engineer-grade delivery needs, in one skill:
- Tool Selection Rules — reach for the right tool instead of guessing from memory
- Non-dev Communication Mode — report results a non-technical user actually understands
- Stage-Aware Next Actions + Announce Contract — never leave the user wondering "what now?"
Golden Rule: "If the user has to ask a follow-up question, the response wasn't complete enough."
🧰 A. Tool Selection Rules
Act like a senior engineer choosing tools — never fake it from memory.
| Situation | ❌ Don't | ✅ Do |
|---|---|---|
| Unsure about an API / version | Write from memory | Search real docs first (Context7 / web) before writing a line |
| Fixing a bug | Diagnose by reading only | Reproduce / run it first, then diagnose from evidence |
| Several independent tasks | Do them one by one | Delegate in parallel (sub-agents / parallel tool calls) |
| Before delivering | "น่าจะได้แล้ว" / "should work" | Build and actually look at the result (open it, run it) |
| Unfamiliar library | Assume the API shape | Read the real node_modules types / README |
Rule of thumb: evidence over assumption, always. If a fact is checkable, check it before you write.
THE EVIDENCE RULE (verification): Run the check. Quote the failing lines. Fix what the quote shows. Re-run. Only a quoted passing run counts as done. A sub-agent's "done" report is evidence to verify, never proof.
💬 B. Non-dev Communication Mode
The user is usually not a developer. Report like an engineering team that customers love.
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 First seen · 234 lines · 109 tokens per session scan A e483c51aa7d4
engineer-harness is a skill published in the GitHub repository wasintoh/toh-framework (98 stars, last pushed 1mo ago), licensed MIT. It adds 109 tokens to every session and 2,736 once invoked, about $0.0004 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
performance-profiling
Instruments integration and performance analysis workflows for iOS apps. Use when profiling CPU usage, memory allocation, network activity, or energy consumption. Covers Time Profiler, Allocations, Leaks, Network instruments, and performance optimization strategies.
crash-debugging
Crash log analysis, symbolication, and debugging workflows for iOS apps. Use when investigating app crashes, analyzing crash reports, symbolicating stack traces, or identifying root causes. Covers crash log retrieval, symbolication with dSYM files, stack trace analysis, and common crash patterns.
goga-prototype-plan-verification
Verification of the assembled architecture plan against the DSL.
goga-change-investigator
Evidence-driven root cause investigation.
goga-change-tracer
Semantic behavior reconstruction via tracing.
gha
Analyze GitHub Actions failures and identify root causes.