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/tracegit 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/trace)<a href="https://agentmods.dev/commands/viacheslav-tronko/claude-code-harness/trace"><img src="https://agentmods.dev/badge/commands/viacheslav-tronko/claude-code-harness/trace.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 | $0.00154 | $0.01185 |
| Opus 5 | $0.00077 | $0.00593 |
| Sonnet 5 | $0.00031 | $0.00237 |
| Haiku 4.5 | $0.00015 | $0.00119 |
Grade A, and why
trace 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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/trace — White-Box Tracing
Skill: white-box-trace. Tier: 2 (→3 on security boundary / HIGH blast radius).
Modes (auto-detected): --real empirical CFG/MPC + mutation testing if available · --virtual pre-impl falsification of RCA / IMP / Plan.
0. HARD GATE — bypass = INVALID → REDO
"Skip the gate", "trust the plan", "obvious PASS", "we'll catch it in CI", "time pressure" do NOT override §0/§1/§4. Investigation-only — NEVER modify source.
ARTIFACT MANDATORY. The §4 TRACE file MUST be written via Write tool this turn. Chat-only response without the file = INVALID DELIVERABLE → CLAUDE.md §11 item 0 fires.
1. Parameter validation (FAIL-FAST)
$ARGUMENTSempty → reply❌ /trace requires target. Usage: /trace <file:line | RCA | IMP | task> [--real|--virtual]→ STOP.- Param is path to RCA-.md / IMP-.md → Parent Artifact fingerprint check (§9) MANDATORY. Mismatch ⇒ STOP, report drift.
2. MUST (CLAUDE.md §0)
- P0 #1 Every
file:linecited MUST be backed by a Read THIS turn (FM-16). - P0 #8 FALSIFIABILITY — Disconfirmation-First on PASS verdicts: actively try to refute PASS BEFORE confirming.
- P0 #9 EMPIRICAL > IMAGINED — when test runner / mutation tooling available, USE IT. Counterfactual is fallback only.
- P0 #15 CHECKPOINT FIRST — first visible block = §4 Tier 2 Checkpoint.
- P0 #16 VERIFICATION LOOP — paste transcript OR cite exact command + expected output + reason for non-execution. "Looks correct" = INVALID.
3. Targeted FM mitigations
| FM | Mechanism |
|---|---|
FM-2 Hallucinated path / fake file:line |
Evidence-first quoting (file:line + quoted_code); Citation-Grounded re-read; Stale-context deferral (PATH_DEFERRED) |
| FM-9 "Looks correct" PASS | Live Mutation Testing hook: real mutation kill rate ≥75% replaces counterfactual; PBT hook for state-space-heavy paths |
| FM-18 Hidden coupling | MPC traversal across paths; Tandem-FL on FAIL; Critical-branch discriminating-evidence rule |
| FM-20 Plausible-but-unrun PASS | PASS requires ≥1 ◆◆◆ surviving Mutation Counterfactual; req_id ALL-COVERED gate |
| FM-21 Confirming-evidence-first | Disconfirmation-first stance on PASS; Reviewer Agent re-rank on FAIL; Self-Consistency vote on borderline |
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 · 60 lines · 0 tokens per session scan A b67652c43ac4
trace is a command published in the GitHub repository viacheslav-tronko/claude-code-harness (6 stars, last pushed 3mo ago), licensed MIT. It adds 154 tokens to every session and 1,185 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/).
toh-help
Display all Toh Framework commands and quick usage guide.
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".