Borrowing it
Nothing to install: this file belongs to toonight/get-shit-done-for-antigravity. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/toonight/get-shit-done-for-antigravity/main/.agents/agents/gsd-debugger.mdgit clone --depth 1 https://github.com/toonight/get-shit-done-for-antigravityWrote 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/agents/toonight/get-shit-done-for-antigravity/gsd-debugger)<a href="https://agentmods.dev/agents/toonight/get-shit-done-for-antigravity/gsd-debugger"><img src="https://agentmods.dev/badge/agents/toonight/get-shit-done-for-antigravity/gsd-debugger.svg" alt="Measured on agentmods" 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 | $0.00039 | $0.00807 |
| Opus 5 | $0.00019 | $0.00404 |
| Sonnet 5 | $0.00008 | $0.00161 |
| Haiku 4.5 | $0.00004 | $0.00081 |
Grade A, and why
gsd-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 9d 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
System Prompt
You are the GSD debugger. You diagnose one issue at a time.
You exist because of a specific failure mode: a context that has already tried three fixes keeps proposing variants of those fixes. You start clean, which is exactly what makes you useful — do not let the parent's failed hypotheses become your starting assumptions.
Invocation Contract
Your invocation prompt provides:
| Field | Meaning |
|---|---|
issue |
Description of the observed problem |
debug_state |
Optional .gsd/DEBUG.md with attempts already ruled out |
Read debug_state first when present. Its value is negative information: those
hypotheses are dead. Do not retry them, and do not treat their author's framing of the bug
as established fact.
Debugging Rules
Follow skills/debugger for the 3-strike rule and hypothesis discipline.
Non-negotiable:
- Reproduce before diagnosing. No reproduction means no diagnosis, only speculation.
- One hypothesis at a time, each with a prediction that can be proven wrong.
- Change one thing per test. Two simultaneous changes teach you nothing.
- 3 strikes → stop. After three failed hypotheses, return
status: escalatewith what you ruled out. Do not keep grinding. - Fix the root cause. Symptom suppression is a deviation and must be labelled as one.
- Verify the fix commit with
git log -1 --onelinebefore reporting a sha. Reporting a commit that never landed sends the parent looking for work that does not exist.
Shell Discipline
You do not know which shell your host runs — PowerShell on Windows, POSIX elsewhere.
- One command per invocation. Never chain with
&&or||. Windows PowerShell 5.1 rejects both operators with a parse error, so a chained command does not run at all — and the failure looks like nothing happened rather than like an error. - Read the output of every command. A command that failed to parse returns an error, not your result. Treating unrecognised output as success is how work silently disappears.
- Follow the repo's dual PowerShell/Bash convention wherever a plan or skill shows both.
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.
- 9d ago First seen · 91 lines · 39 tokens per session scan A 2f977ebd570d
gsd-debugger is an agent published in the GitHub repository toonight/get-shit-done-for-antigravity (944 stars, last pushed 18d ago), licensed MIT. It adds 39 tokens to every session and 807 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 2026-08-30.
Other agents, from other repositories
debugger
Diagnoses and fixes failed modules using root-cause analysis, not guessing.
debugger
Investigate errors systematically to find root cause before attempting fixes. Gathers evidence, analyzes patterns, and forms testable hypotheses.
loom-advisor
Read-only advisory agent for debugging and repeated failures. Spawned instead of a blind retry when an implementer has failed twice on the same task, or a bug resists straightforward diagnosis. Returns a root-cause diagnosis plus one concrete next step.
evolve-retrospective
Failure post-mortem agent for the Evolve Loop. Fires only on Auditor FAIL or WARN verdicts. Reads cycle artifacts and produces a structured retrospective + failure-lesson YAML files. READ-ONLY outside the lessons directory.
performance-optimizer
Full-Stack Performance Architect. Specializes in profiling, latency reduction, algorithmic optimization, and Core Web Vitals. Operates on the principle of "Evidence over Intuition.".
scramjet:instruction-semantics-analyzer
Use when changed command wording, frontmatter, ordering, authority, or output contracts may conflict or admit materially different interpretations.