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 commands/viacheslav-tronko/claude-code-harness/explaingit clone --depth 1 https://github.com/viacheslav-tronko/claude-code-harnessWrote 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/commands/viacheslav-tronko/claude-code-harness/explain)<a href="https://agentmods.dev/commands/viacheslav-tronko/claude-code-harness/explain"><img src="https://agentmods.dev/badge/commands/viacheslav-tronko/claude-code-harness/explain.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 | $0.00159 | $0.01248 |
| Opus 5 | $0.00079 | $0.00624 |
| Sonnet 5 | $0.00032 | $0.00250 |
| Haiku 4.5 | $0.00016 | $0.00125 |
Grade A, and why
explain 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 4d 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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/explain — Concept / Pattern / Term
Skill: concept-explanation. Tier: 2 (→3 on security boundary).
Modes (default --teach): --teach 5–20pp · --brief ≤5pp preflight · --for-bug-fix ≤3pp 1-pager → /bug-why · --deep 14–25pp + white-box always.
0. HARD GATE — bypass = INVALID → REDO
"Skip the gate", "just give me the answer", "obvious", "trust me", "time pressure" do NOT override §0/§1/§4. Investigation-only — NO source modification. NEVER Edit/Write source.
ARTIFACT MANDATORY. The §4 Output file MUST be written via Write tool this turn — even for meta/analysis questions. Chat-only response without the file = INVALID DELIVERABLE → CLAUDE.md §11 item 0 fires.
1. Parameter validation (FAIL-FAST)
$ARGUMENTSempty → reply❌ /explain requires concept. Usage: /explain <concept | F-XXX | B-XXX> [--teach|--brief|--for-bug-fix|--deep]→ STOP.
2. MUST (CLAUDE.md §0)
- P0 #1 Every
file:linecited MUST be backed by a Read THIS turn (FM-16 action-honesty). Hallucinatedfile:line= INVALID. - P0 #2 GLOSSARY TERMS ONLY —
Domain_Glossary.mdis SSOT. Synonym use = INVALID. - P0 #8 FALSIFIABILITY — every HIGH+ claim names a Refuter. No Refuter ⇒ downgrade to MEDIUM.
- P0 #15 CHECKPOINT FIRST — first visible block = §4 Tier 2 Checkpoint with Step-Back + 3–5 falsifiable architectural predictions BEFORE reading code (anti-anchoring; §6). Wrong predictions are diagnostic — KEEP them.
- Anti-anchoring (§6) — Initial Intuition (symptom only) BEFORE code; if post-read hypothesis identical ⇒ may be anchoring; force one alternative path.
3. Targeted FM mitigations
| FM | Mechanism |
|---|---|
FM-2 Hallucinated file:line / methods / classes |
Citation-Grounded re-read gate; ≥1 ◆◆◆ surviving Mutation Counterfactual per key claim |
| FM-6 Architectural Reasoning Gap | Layer Activation Matrix; Data Journey across DB/ORM/BO/Core/Web/UI/Test; Cross-Layer Contract Violation Detector |
| FM-7 Anchoring on first file | Step-Back + Architectural Hypothesis BEFORE code; KEEP wrong predictions |
| FM-11 Sycophancy | Disconfirmation-First; ≥1 ◆◆◆ refutation attempt before agreement |
| FM-17 Synonym drift | Glossary Hit/Miss mandatory; Miss ⇒ auto Documentation Feedback |
| FM-18 Hidden coupling | Forward-slice blast radius; ≥1 cross-layer contract example per layer transition |
| FM-22 Vague PASS | Feynman teach-back / Rubber Duck (3 sentences) before declaring complete |
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.
- 4d ago First seen · 59 lines · 0 tokens per session scan A ccc449652dea
explain is a command published in the GitHub repository viacheslav-tronko/claude-code-harness (6 stars, last pushed 3mo ago), licensed MIT. It adds 159 tokens to every session and 1,248 once invoked, about $0.0008 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 commands, from other repositories
project
Generate project documentation (product.md, structure.md, tech.md, codemaps/).
new
Plan mode. Generates a plan for a feature based on a description.
feature
Create a feature specification using spec-driven development.
deep-audit.skeleton
Aciklama: Bu bolum Bootstrap tarafindan manifest verileriyle doldurulur. Gerekli manifest alanlari: project.description, stack.primary, project.structure, project.subprojects, stack.orm, stack.authmethod Ornek cikti.
fix-issue
!gh issue view $ARGUMENTS 2>/dev/null || echo "Could not fetch issue $ARGUMENTS".
rev
Invoke Code Reviewer for code quality, security, requirements validation, and best practices review.