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 skills/blake-simpson/belmont/debugnpx skills add blake-simpson/belmont --skill debuggit clone --depth 1 https://github.com/blake-simpson/belmontWhat 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 | $0.00033 | $0.00628 |
| Opus 5 | $0.00016 | $0.00314 |
| Sonnet 5 | $0.00007 | $0.00126 |
| Haiku 4.5 | $0.00003 | $0.00063 |
Grade A, and why
debug 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 2d 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Belmont: Debug
This is the debug router. It directs you to the appropriate debug sub-workflow.
Two Modes
| Mode | Skill | Verification | Best for |
|---|---|---|---|
| Auto | /belmont:debug-auto |
Verification agent checks each fix | Complex logic bugs, race conditions, issues needing automated testing, narrow code-only fixes |
| Manual | /belmont:debug-manual |
User checks each fix (adds debug logs); deep Belmont context + in-place spec reconciliation after the fix | UI bugs, visual issues, known repro steps, bugs that exist because the spec drifted from reality, multi-feature debugging |
The two modes diverge sharply after the fix is confirmed:
- Auto commits the code change and ends. The PRD, TECH_PLAN, NOTES stay as they are.
- Manual runs a Spec Reconciliation phase that walks the loaded specs, identifies drift, presents diffs for user approval, and commits code + spec edits atomically. This is how Belmont's memory gets corrected so future sessions don't operate on stale truth.
Route Decision
Check the user's invocation text for mode hints:
Route to /belmont:debug-auto if the user mentions:
- "auto", "automatic", "full verification", "run tests", "agent verify"
- "just the code", "narrow fix", "don't touch specs"
- Complex or hard-to-reproduce issues where agent-driven iteration beats user-in-the-loop
Route to /belmont:debug-manual if the user mentions:
- "manual", "I'll check", "I'll verify", "I can test", "quick", "fast"
- "debug logs", "console.log", "logging"
- UI bugs, visual issues, styling problems
- "I know how to reproduce"
- "spec is wrong", "PRD is stale", "TECH_PLAN doesn't match", "fix the docs too", "reconcile specs"
- "across two features", "multi-feature", "spans X and Y"
If unclear, ask the user:
Which debug mode?
- Auto (
/belmont:debug-auto) — dispatches a verification agent to check each fix. Narrow code-only fix. More thorough but slower.- Manual (
/belmont:debug-manual) — you verify each fix yourself. Loads deep Belmont context (master PR_FAQ + master/feature PRD + TECH_PLAN + NOTES + latest MILESTONE.done). After the fix, walks the loaded specs and offers diffs to reconcile drift in place. Best for bugs that may have a spec-level root cause.
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.
- 2d ago First seen · 47 lines · 33 tokens per session scan A aef8aba92ddc
debug is a skill published in the GitHub repository blake-simpson/belmont (4 stars, last pushed 14d ago), licensed Apache-2.0. It adds 33 tokens to every session and 628 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.
Other skills, from other repositories
agent-performance-benchmarker
Agent skill for performance-benchmarker - invoke with $agent-performance-benchmarker.
agent-quorum-manager
Agent skill for quorum-manager - invoke with $agent-quorum-manager.
agent-agent
Agent skill for agent - invoke with $agent-agent.
agent-crdt-synchronizer
Agent skill for crdt-synchronizer - invoke with $agent-crdt-synchronizer.
agent-adaptive-coordinator
Agent skill for adaptive-coordinator - invoke with $agent-adaptive-coordinator.
agent-benchmark-suite
Agent skill for benchmark-suite - invoke with $agent-benchmark-suite.