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.
git clone --depth 1 https://github.com/Langerrr/zforgeWrote 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/langerrr/zforge/debug)<a href="https://agentmods.dev/commands/langerrr/zforge/debug"><img src="https://agentmods.dev/badge/commands/langerrr/zforge/debug/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.
<a href="https://agentmods.dev/commands/langerrr/zforge/debug"><img src="https://agentmods.dev/badge/commands/langerrr/zforge/debug.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00008 | $0.00554 |
| Opus 5.5 | $0.00003 | $0.00222 |
| Sonnet 5.5 | $0.00002 | $0.00111 |
| Haiku 4.5 | $0.00001 | $0.00055 |
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 1mo 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.
/debug — Debugging Router
A concrete failure — a broken test, a reproducible bug, a failing deploy. This command owns no debugging opinion of its own; it names which methodology applies and in what order, because the failure mode is applying one of them and skipping the rest.
Not for open-ended exploration ("this feels wrong"). That is ce:ideate territory.
Sequence
1. superpowers:systematic-debugging — always start here.
Investigation discipline before any fix. This is the one step that is never skipped, including when the cause seems obvious. Especially then.
2. compound-engineering:ce-debug — reproduce, trace, predict.
Its hypothesis-and-prediction gate goes before any code change: state what you believe is wrong and what you expect to see if you are right, then check. A fix applied without a failed prediction is a guess that happened to be typed confidently.
3. zforge:async-reasoning — add when the failure involves timing.
Async state, race conditions, cache invalidation, init order, stale reads, write-then-read gaps within one component.
4. distributed-architect:dist-debug — add when the failure crosses a boundary.
Cross-service, cross-process, distributed races. Backward tracing from the symptom to the boundary where the cause lives. Check the project's topology file first if it has one.
While debugging a planned feature
If the failure is in a feature with a zforge directory, the ledger is evidence:
decision_review.mdmay already contain the decision that caused this.- An open standing flag may already name this defect class — a bug in the gap a flag describes is a flag being discharged the expensive way.
- The phase's
## Evidence Requiredshows what was never verified, which is usually where to look first.
After the fix
Record it where it will be found again:
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.
- 1mo ago First seen · 47 lines · 8 tokens per session scan A 11ef9d6eab30
debug is a command published in the GitHub repository Langerrr/zforge (10 stars, last pushed 1mo ago), licensed MIT. It adds 8 tokens to every session and 554 once invoked, about $0.0000 per session on Opus 5.5. 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
crystallize
After a session or an incident, turn what happened into reusable knowledge — learn captures this session's lessons, review/approve promotes incident patterns into agent improvements, skill turns a repeating procedure into a skill.
debug
Systematic debugging: root cause before any fix, one hypothesis at a time, and a three-strikes rule that questions the architecture instead of stacking patches.
recover
Diagnose and recover from interrupted or corrupted state.
analyze-misfires
Identify skills injected where not needed, propose regex and description tightening.
debug
Investigate bugs and failures with systematic root-cause analysis.
bugfix
TDD-driven bugfix workflow: tester writes failing test (RED) → developer fixes (GREEN) → developer refactors (REFACTOR) → reviewer validates. Accepts issue number, description, or both. Asks before pushing and creating a PR; --no-pr commits locally without asking.