bug-protocol

bug-protocol is a skill for Claude Code, Codex from austin-starks/Public-Portfolio-Challenge. It costs 103 tokens per session (890 once invoked), scanned A, original, MIT.

A procedure for reporting failures during NexusTrade certification campaigns, where trading configurations and results are tested. It requires failures to be reproduced, recorded, and separated from results that may be affected by bugs.

In plain words
What is it for?
Investigating tool errors, hangs, unexpected numbers, or behavior that conflicts with the configured trading campaign, then documenting what the issue blocks.
Why use it?
It prevents silent workarounds and unsupported explanations from hiding configuration, data, or engine problems.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Investigating tool errors, hangs, unexpected numbers, or behavior that conflicts with the configured trading campaign, then documenting what the issue blocks.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/austin-starks/public-portfolio-challenge/bug-protocol
Install

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.

Any agent
npx skills add austin-starks/Public-Portfolio-Challenge --skill bug-protocol
Clone the repo
git clone --depth 1 https://github.com/austin-starks/Public-Portfolio-Challenge

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for bug-protocol

README.md
[![agentmods](https://agentmods.dev/badge/skills/austin-starks/public-portfolio-challenge/bug-protocol/github.svg)](https://agentmods.dev/skills/austin-starks/public-portfolio-challenge/bug-protocol)
Your own site
<a href="https://agentmods.dev/skills/austin-starks/public-portfolio-challenge/bug-protocol"><img src="https://agentmods.dev/badge/skills/austin-starks/public-portfolio-challenge/bug-protocol/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.

agentmods 80×15 button for bug-protocol

Your own site · 80×15
<a href="https://agentmods.dev/skills/austin-starks/public-portfolio-challenge/bug-protocol"><img src="https://agentmods.dev/badge/skills/austin-starks/public-portfolio-challenge/bug-protocol.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 103 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 890 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00103 $0.00890
Opus 5 $0.00051 $0.00445
Sonnet 5 $0.00021 $0.00178
Haiku 4.5 $0.00010 $0.00089

Measured 9d ago against content hash 32c8e5400a1f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

bug-protocol 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 9d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (install.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

skills/bug-protocol/SKILL.md · 60 lines

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.

Bug Protocol — "Loudly Declare"

The compute + engine stack is new and moves. Every failure gets this protocol. Never paper over an error to keep a campaign moving — a swallowed bug is worse than a blocked stage. A confabulated mechanism (inventing why the engine did something) is the worst failure mode: the campaign's worst moments have been confabulated reconcile events, pricing behavior, and budget behavior. Verify, don't assert.

Precedent: a totalBudget over-deployment bug (a book with a 40% budget deployed ~96% of NAV at cold start) was caught exactly this way and invalidated weeks of results. The general lesson is to quarantine every result affected by an engine defect and rerun only after the behavior is corrected.

The five steps (apply on every failure)

  1. Stop and characterize — is it config (my fault), data (source-side), or engine (NexusTrade)? Reproduce minimally — the smallest job/backtest that still shows it.
  2. Log it immediately in the run's CAMPAIGN_LOG.md under a ⚠️ BUG/ISSUE heading: symptom, repro IDs, expected-vs-observed, current hypothesis (labeled hypothesis), what it blocks.
  3. If it's a real engine/data bug, write a hand-off doc <NAME>_BUG.md in the run directory, modeled on the template in references/BUG_TEMPLATE.md (one-line summary, expected vs observed, reproduction with IDs, leading hypothesis, alternatives to rule out, where to look, impact on results).
  4. Quarantine affected results — mark every backtest/study that ran before the fix as suspect/invalid. Re-run after the fix. If the bug inflated leverage, both returns AND drawdowns are inflated and any sweep gene it touched may be non-differentiating — re-validate all of it.
  5. Tell the human, prominently, in the chat. Bugs are a co-equal deliverable of the episode with the trading result — not an embarrassment to route around.

What is NOT a bug (honest-signal failures — don't file these)

  • A fold returning NoEligibleCandidate = an honest signal, not a bug. Widen validation; don't drop the certification floors.
  • A dead / not-yet-listed name (e.g. SNDK in 2022) being absent or in a reject list with no fabricated values = correct behavior. The engine fabricating a price/indicator value for a name that didn't trade IS a bug (see engine-sanity).
  • A single-tick reconcile target = the live strategy accretes the rest over daily rebalances — say so, don't re-diagnose it as a bug (see deploy-gate).

Read the full file on GitHub · 60 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 9d ago First seen · 60 lines · 103 tokens per session scan A 32c8e5400a1f

Subscribe to this mod's changes

bug-protocol is a skill published in the GitHub repository austin-starks/Public-Portfolio-Challenge (43 stars, last pushed 2d ago), licensed MIT. It adds 103 tokens to every session and 890 once invoked, about $0.0005 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 2026-08-30.

Related

Other skills, from other repositories