Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/zxpmail/ReqForgenpx agentmods add skills/zxpmail/reqforge/bug-fixerWrote 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/skills/zxpmail/reqforge/bug-fixer)<a href="https://agentmods.dev/skills/zxpmail/reqforge/bug-fixer"><img src="https://agentmods.dev/badge/skills/zxpmail/reqforge/bug-fixer/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/skills/zxpmail/reqforge/bug-fixer"><img src="https://agentmods.dev/badge/skills/zxpmail/reqforge/bug-fixer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 6 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Prompt Injection · line 1 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Prompt Injection · line 15 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Prompt Injection · line 50 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Prompt Injection · line 50 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- medium Excessive Agency · line 34 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 148 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00000 | $0.02456 |
| Opus 5.5 | $0.00000 | $0.00982 |
| Sonnet 5.5 | $0.00000 | $0.00491 |
| Haiku 4.5 | $0.00000 | $0.00246 |
Grade A, and why
bug-fixer 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 10d 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 — 179 lines — stays where its author put it; the contents beside it link to each section on GitHub.
name: bug-fixer description: Used when the user says "this feature is broken", "getting an error", "something's not right", or reports a bug, compilation error, or runtime exception. Locates root cause through a four-stage systematic debugging process and fixes it. version: 1.2.0 updated: 2026-06-27 requires: []
[Task] Locate the root cause of bugs through a systematic debugging process and fix them. Fix one problem at a time. Assess impact before each modification. Verify regression after fix.
[Invocation Context] bug-fixer may be called in two scenarios: 1. User directly reports a bug -> main Agent invokes bug-fixer -> after fix, suggest user run /code-review to verify 2. code-review finds confirmed bug/security/type issues (confidence_5 ≥ 4, or legacy confidence ≥ 0.6) -> main Agent invokes bug-fixer, passing the failure items from the code-review report -> after fix, main Agent re-dispatches code-review
**Action filter**: bug-fixer receives **only `auto-fix` findings**. Findings with `action="ask-user"` (intent / product-behavior / dead-code decisions) are escalated to the human by dev-builder *before* reaching bug-fixer — never auto-fix them. See `../_shared/finding-actions.md`.
[Not For] - Feature requests or new functionality -> use /dev-builder instead - Code quality or style issues without runtime errors -> use /code-review instead - Performance optimization without a specific bug -> use /code-review with performance dimension
[Dependency Check] Automatically executed as the first step when the Skill starts:
Required:
- Project code exists -> if no code, prompt to call /dev-builder first
- Bug description -> user-provided symptoms, or failure item descriptions from a code-review report
Optional (enhances debugging capability):
- Product-Spec.md -> if available, cross-reference expected behavior to determine if it is a bug or a feature
- DEV-PLAN.md -> if available, locate the relevant Phase and files
- Design tool MCP (Pencil / Figma, etc.) -> if available, cross-reference design to check if UI is correct
- Playwright plugin -> if available, automate reproduction and verification
- git -> if available, use git log/diff/blame to trace changes
- **Dependency Graph** (`dep-graph`) -> if available, run `pnpm dep-graph affected <file>` to scope the blast radius before debugging
- **forge-bug-fix** (`pnpm forge-bug-fix`) -> if available, use diagnose/trace/verify to automate preflight checks, capture debug snapshots, and run post-fix verification
[Shared Discipline]
Karpathy 四原则 → ../_shared/karpathy-discipline.md（bug 场景：先证据后改码；最小修复）
只修 auto-fix finding；ask-user（意图/产品行为/死代码决策）已被上游 escalate → ../_shared/finding-actions.md
[First Principles]
Debug 前必读 references/first-principles.md
[Output Style]
→ references/output-style.md
→ Bug 报告 / 修复完成必须附加 ../_shared/output-status-protocol.md（Status: BLOCKED 或 NEEDS_CONTEXT 时必须说明原因）
[File Structure]
bug-fixer/ ├── SKILL.md ├── commands/bug-fixer.md └── references/ ├── first-principles.md ├── output-style.md ├── debugging-strategy.md # 四阶段（Stage 1–4） ├── cot-diagnostic-checklist.md ├── three-layer-diagnostic-model.md ├── debugging-rule-checklist.md ├── anti-rationalization.md ├── anti-ai-slop-checklist.md # 交付前自检：防敷衍式修 bug ├── workflow.md # Startup → Debug → Verify → Complete └── yolo-mode.md ../_shared/
What ships with it
13 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- commands/bug-fixer.md 572 B
- references/anti-ai-slop-checklist.md 3.4 KB
- references/anti-rationalization.md 1.2 KB
- references/cot-diagnostic-checklist.md 560 B
- references/debugging-rule-checklist.md 3.4 KB
- references/debugging-strategy.md 4.5 KB
- references/dimension-checklist.md 3.3 KB
- references/first-principles.md 1.9 KB
- references/output-style.md 1.5 KB
- references/three-layer-diagnostic-model.md 1.1 KB
- references/workflow.md 5.7 KB
- references/yolo-mode.md 452 B
- skill.json 924 B
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.
- 10d ago Changed 34fda58f2606
- 1mo ago First seen · 179 lines · 0 tokens per session scan A 34325c75bd0d
bug-fixer is a skill published in the GitHub repository zxpmail/ReqForge (18 stars, last pushed 10d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,456 tokens. 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 skills, from other repositories
systematic-debugging
A step-by-step method for finding the underlying cause of technical problems before changing code. It covers reading errors, reproducing failures, checking recent changes, and tracing data across system components.
openrig-skills
Use when an agent running inside OpenRig needs the right skill or context for fleet recovery, seat handover, new-seat orientation, a watchdog wake, cross-host reach to an agent on another machine, rig packaging, an OpenRig upgrade, systematic debugging, queue triage, or implementation planning; also when the agent…
diagnosing-superpowers
A diagnostic workflow for investigating problems in Superpowers agent sessions using saved conversation records. It requires evidence linked to exact file paths and line numbers, then organizes findings into a report.
systemic-issue-triage
Trigger: new issue, bug report, triage, backlog, issue flood, community report, root cause, dead-end, blocked user. Attack issues by root class, never one-by-one; fixes must shrink the system, not grow it.
issue-root-resolution
Trigger: root audit, atacar la raíz, issue roots, backlog roots, mechanism map, deletion-driven fix, resolver issues de raíz, close outdated issues. Audit and resolve issue clusters by verified root cause.
oversight-team
Use when you are a seat on the oversight pod (a standing monitor-mode rig that keeps OTHER rigs healthy), configuring or running the drift detectors, or choosing whether to intervene vs escalate. Covers the pull-not-poll posture, the v0 detectors (premature-park, process-drift, off-task, token-burn), the intervention…