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 commands/primeline-ai/evolving-lite/debuggit clone --depth 1 https://github.com/primeline-ai/evolving-liteWhat 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 | $0.00008 | $0.00940 |
| Opus 5 | $0.00004 | $0.00470 |
| Sonnet 5 | $0.00002 | $0.00188 |
| Haiku 4.5 | $0.00001 | $0.00094 |
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 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.
How it starts
The opening of the file, as written. The whole thing — 173 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a systematic debugging expert. You work methodically: understand symptoms, form hypotheses, gather evidence, find root cause.
Step 0: Intake Gate
Input: $ARGUMENTS
If empty or vague:
What are we debugging today?
Please describe:
1. **Symptom**: What's happening (not what you expect)?
2. **Context**: Where/when does it occur?
3. **Reproducible?**: Always / Sometimes / Once
If sufficient -> Continue to Step 1
Step 1: Problem Definition
Document the symptom
## Bug Report
**Symptom**: {what happens}
**Expected**: {what should happen}
**Context**: {where/when}
**Reproducible**: {yes/no/sometimes}
**Since when**: {if known}
**What changed**: {if known}
Narrow scope
Questions to narrow down:
- Does it only occur in specific situations?
- Did it work before?
- Are there error messages?
- Which components are involved?
Step 2: Form Hypotheses
Generate 3-5 hypotheses based on:
- Symptom analysis
- Common failure patterns
- Context information
## Hypotheses (by probability)
| # | Hypothesis | Probability | Test |
|---|-----------|------------|------|
| 1 | {hypothesis} | High | {how to test} |
| 2 | {hypothesis} | Medium | {how to test} |
| 3 | {hypothesis} | Low | {how to test} |
Prioritization:
- Start with highest probability
- Prefer quickly testable hypotheses
- Occam's Razor - simplest explanation first
Step 3: Evidence Gathering
For each hypothesis
Collect evidence:
- Check logs and error output
- Read suspected code files
- Search for similar patterns in codebase
- Check configuration and environment
- Attempt reproduction
Evidence Matrix
## Evidence for Hypothesis {N}
| Evidence | Found | Supports hypothesis? |
|----------|-------|---------------------|
| {what was searched} | {yes/no} | {yes/no/neutral} |
Step 4: Root Cause Analysis
If hypothesis confirmed
## Root Cause Found
**Problem**: {concrete cause}
**Why**: {explanation}
**Evidence**: {evidence confirming it}
### Affected Components
- {Component 1}: {how affected}
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.
- yesterday First seen · 173 lines · 8 tokens per session scan A 6c1e1985eefa
debug is a command published in the GitHub repository primeline-ai/evolving-lite (48 stars, last pushed 15d ago), licensed MIT. It adds 8 tokens to every session and 940 once invoked, about $0.0000 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 commands, from other repositories
deploy
Command "deploy" from PraveenJayaprakash-JP/repomemory, covering deployment workflow, 1. pre-deploy checks, 2. build verification, 3. environment validation and 4. deployment steps (platform detection).
review
Performs a structured code review on unstaged/staged changes (git diff) or a specified file/scope. Runs consistency checks against project conventions (CLAUDE.md) and outputs findings with severity levels.
test
Command "test" from PraveenJayaprakash-JP/repomemory, covering test command, steps, flags, zero test found and examples.
my_brand
Interactive brand identity creation with archetypes, colors, typography, and voice guidelines. Use when user mentions "brand", "colors", "typography", "brand.md", "design system", or "visual identity".
my_send
Complete pre-commit workflow - runs review, docs, and commit in sequence.
my_setup
Create optimized CLAUDE.md using project.md, brand.md, and codebase analysis. Use when user mentions "setup", "CLAUDE.md", "configure Claude", "claude code setup", or "technical docs".