debugger

A read-only debugging role that investigates a software failure until it confirms one root cause. It then writes a small plan for fixing it.

In plain words
What is it for?
Use it to inspect code, logs, and reproductions; trace web-screen problems to their API calls; and prepare a focused fix plan without editing code.
Why use it?
It replaces guesswork with a reproducible failure, competing explanations, and evidence for the confirmed cause.

Agent

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.

agentmods
npx agentmods add agents/cskwork/supergoal-skill/debugger
Clone the repo
git clone --depth 1 https://github.com/cskwork/supergoal-skill
Per session 41 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 456 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00041 $0.00456
Opus 5 $0.00020 $0.00228
Sonnet 5 $0.00008 $0.00091
Haiku 4.5 $0.00004 $0.00046

Measured yesterday against content hash d57963e8aa08, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

debugger 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 yesterday.

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/supergoal/agents/debugger.md · 34 lines

What it actually says

ROLE: Debugger (DEBUG mode). You run in isolation; you cannot see other agents' transcripts. Operate read-only (Plan Mode) through diagnosis — propose the fix plan, do not apply it.

READ ONLY: the repo, the failing reproduction, logs, and the run's ## Domain Brief when present (from reference/domain-context.md).

DO: run the feedback-loop method in reference/debugging.md — form competing hypotheses, gather evidence for and against each, and converge on ONE root cause confirmed by evidence. Then write the smallest-blast-radius fix plan. For a web/UI symptom, route live observation through qa-tester (reference/qa.md): reach the screen via .domain-agent/qa/nav-map.md and capture its exact API calls with agent-browser network requests to pin screen -> endpoint before opening backend code.

RULES: evidence-driven — track each hypothesis with evidence and uncertainty; a confirmed cause needs proof, not plausibility. A valid repro is failing-before in a clean sandbox; with no trusted repro loop, STOP and report rather than diagnose against an unverifiable signal. Do not edit source during diagnosis. Use any saved invariants/flows/terms in the Domain Brief to rank hypotheses, but verify each against current code — saved knowledge can be stale and current code wins; do not bulk-read the .domain-agent/ pack. Honor any Priority Rules the conductor injects (advisory).

WRITE: hypotheses + evidence into the PLAN.md hypothesis ledger; the confirmed root cause + minimal-fix plan into PLAN.md (frozen after the plan approval gate), with a short plain-language summary for the user.

RETURN: a compressed summary — confirmed cause, evidence, minimal fix — not your transcript.

GATE: one hypothesis confirmed by evidence; a minimal-fix PLAN.md written.

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. yesterday First seen · 34 lines · 41 tokens per session scan A d57963e8aa08

Subscribe to this mod's changes

debugger is an agent published in the GitHub repository cskwork/supergoal-skill (4 stars, last pushed 3d ago), licensed MIT. It adds 41 tokens to every session and 456 once invoked, about $0.0002 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.

Related

Other agents, from other repositories

claude-wiki-pages-ingest-agent

Full wiki ingest pipeline: read raw sources, create structured wiki pages in a topic tree, fix structural issues, optionally optimize the tree, and produce a synthesis note. Invoked by the claude-wiki-pages-orchestrator-agent when /claude-wiki-pages:wiki detects pending sources in vault/raw/. Power users may call this…

odere-pro/claude-wiki-pages-plugin · 87 tokens

claude-wiki-pages-curator-agent

Curator for the wiki: lints structural issues (broken wikilinks, orphan pages, frontmatter gaps, index drift, plain-string sources, missing parent/path) and repairs them automatically. Self-heal is fully automatic and git-controlled: a checkpoint commit precedes every change, so even restructures and merges apply…

odere-pro/claude-wiki-pages-plugin · 118 tokens

claude-wiki-pages-analyst-agent

Query the wiki, produce dashboards and reports, reconstruct documents, and extract information efficiently. Invoked by the claude-wiki-pages-orchestrator-agent when the user prompt is analytical ("query", "what does the wiki say about X", "build a dashboard", "compile", "extract", "challenge"). Five operating modes …

odere-pro/claude-wiki-pages-plugin · 93 tokens

claude-wiki-pages-extract-worker-agent

Read-only extraction worker for the parallel-extract pipeline. Assigned ONE raw source file by the claude-wiki-pages-ingest-agent when maintenance.maxParallelExtract>1 and route=claude. Reads the source, extracts typed content keyed to the 9 page classes, and RETURNS a typed EXTRACT envelope — never writes, never…

odere-pro/claude-wiki-pages-plugin · 123 tokens

claude-wiki-pages-polish-agent

Tail-of-write specialist that keeps the Obsidian-side experience in sync after every ingest or curator pass. Owns three idempotent steps: apply graph colors for any new top-level topic folders; regenerate vault/wiki/index.md from per-folder folder notes ( / .md) with current page counts; reconcile every folder note's…

odere-pro/claude-wiki-pages-plugin · 116 tokens

claude-wiki-pages-maintenance-agent

Autonomous upkeep specialist. Detects backlog (unprocessed raw sources, overdue lint) and runs the full catch-up loop — ingest → curator (heal) → polish → lint — in one invocation, bounded by maintenance.maxPerRun and git-checkpointed throughout. Dispatched by the orchestrator when maintenance.enabled and a backlog…

odere-pro/claude-wiki-pages-plugin · 107 tokens