debugger
Diagnose and, when explicitly requested, narrowly fix LogOS vNext build, test, boot, UEFI, QEMU, scheduler, or proof failures using reproduction and evidence.
Skill by Eresea
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/eresea-debugger
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: testing
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/debugger
What's inside
- Skill name
debugger- Description
- Diagnose and, when explicitly requested, narrowly fix LogOS vNext build, test, boot, UEFI, QEMU, scheduler, or proof failures using reproduction and evidence. Use for a specific failure or regression; do not use for broad feature planning or general code review.
- License
- MIT
- Metadata
- short-description: Diagnose a LogOS failure
- Instructions
- 40 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- files: referenced but missing: scripts/check.ps1, scripts/run.ps1
Files (1)
- SKILL.md
SKILL.md (first lines)
# LogOS debugger Find the smallest evidence-backed root cause of a specific LogOS failure. Diagnose before editing. Apply a fix only when the request explicitly asks for one or the debugging task clearly includes remediation; otherwise stop after the diagnosis and recommended patch. ## Establish the failure 1. Read `AGENTS.md`, `docs/README.md`, and the active docs for the affected subsystem. 2. Capture the exact command or boot path, expected result, actual result, environment, repeatability, and the last known-good revision or diff when available. 3. Inspect `git status --short` and the current diff before changing anything. Preserve unrelated user work. 4. Reproduce with the smallest relevant command. Use host tests for pure state logic; use `scripts/check.ps1` or `scripts/run.ps1` for UEFI, QEMU, boot, and hardware-facing failures. 5. For boot failures, preserve and inspect QEMU debug-console output. Do not hide a failure by weakening assertions, suppressing output, increasing arbitrary timeouts, or changing proof workloads without evidence. ## Narrow the cause - Classify the failure as build/dependency, host-test logic, UEFI handoff, per-CPU/SMP, scheduler/preemption, memory/bounds, proof integration, or documentation/tooling. - Form a small set of falsifiable hypotheses, then test them with targeted inspection or one focused experiment at a time. - Trace the failing value or state transition to its owner. Check invariants at the boundary where the bad state is introduced, not only where it becomes visible. - Prefer existing bounded types, scheduler transitions, proof hooks, and diagnostics. Do not introduce a new subsystem or dependency to make a diagnosis easier. - For concurrency or boot failures, distinguish a deterministic logic error from timing sensitivity and record the evidence for that distinction. ## If remediation is in scope - Make the smallest patch at the responsible layer. - Preserve `no_std`, fixed-size resource bounds, independently bootable increments, and QEMU debug-console output. - Add a focused regression test or proof assertion when practical. - Follow the project verification rules after Rust changes: `cargo fmt --check`, the smallest relevant host test, and `cargo clippy -- -D warnings`. Run the relevant boot or UEFI script as well when applicable. - Do not bundle refactors, speculative hardening, or unrelated cleanup. ## Deliverable Return: 1. **Reproduction** — exact command/path and observed failure. 2. **Root cause** — the responsible file, symbol, state transition, and violated invariant. 3. **Evidence** — checks or experiments that ruled out competing explanations. 4. **Fix** — changed files and why the patch is sufficient, if remediation was requested. 5. **Verification** — commands and exact results, including anything blocked. 6. **Remaining risk** — only concrete unresolved behavior or follow-up.
From Eresea/LogOS, MIT.
More from Eresea/LogOS
- 0.520score
implementer
Eresea
Implement an approved or clearly scoped LogOS vNext change with a minimal independently bootable patch and proportional verification.
SkillBFreshness: BClaude Code +9 - 0.517score
planner
Eresea
Plan bounded LogOS vNext changes before implementation, including affected ownership, documentation, ADR, and verification work.
SkillBFreshness: BClaude Code +9 - 0.499score
domain-modeling
Eresea
Build and sharpen a project's domain model.
SkillBFreshness: BClaude Code +9 - 0.499score
codebase-design
Eresea
Shared vocabulary for designing deep modules.
SkillBFreshness: BClaude Code +9 - 0.499score
grilling
Eresea
Grill the user relentlessly about a plan, decision, or idea.
SkillBFreshness: BClaude Code +9 - 0.492score
pr
Eresea
Use when writing a PR body.
SkillBFreshness: BClaude Code +9 - 0.488score
implement
Eresea
Implement a piece of work based on a spec or set of tickets.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 14
- Last commit (this folder)
- Aug 18, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/Eresea/LogOS
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Eresea
- 0.520score
implementer
Eresea
Implement an approved or clearly scoped LogOS vNext change with a minimal independently bootable patch and proportional verification.
SkillBFreshness: BClaude Code +9 - 0.517score
planner
Eresea
Plan bounded LogOS vNext changes before implementation, including affected ownership, documentation, ADR, and verification work.
SkillBFreshness: BClaude Code +9 - 0.499score
domain-modeling
Eresea
Build and sharpen a project's domain model.
SkillBFreshness: BClaude Code +9 - 0.499score
codebase-design
Eresea
Shared vocabulary for designing deep modules.
SkillBFreshness: BClaude Code +9 - 0.499score
grilling
Eresea
Grill the user relentlessly about a plan, decision, or idea.
SkillBFreshness: BClaude Code +9 - 0.492score
pr
Eresea
Use when writing a PR body.
SkillBFreshness: BClaude Code +9
Cite this page
debugger skill. Agent Mart. https://agentmart.dev/skills/eresea-debugger (last updated 2026-10-24)