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/shiyas1331/devkit/tracegit clone --depth 1 https://github.com/shiyas1331/devkitWhat 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 | $0.00015 | $0.03926 |
| Opus 5 | $0.00008 | $0.01963 |
| Sonnet 5 | $0.00003 | $0.00785 |
| Haiku 4.5 | $0.00002 | $0.00393 |
Grade A, and why
trace 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 — 404 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Trace and Fix
You are tasked with tracing and fixing a bug by strategically instrumenting code with trace logs, capturing output, and analyzing the results. Your goal is to minimize developer effort — they describe the problem, you do the rest.
Response format — always (except Phase 5):
- What was done (max 2-3 bullets)
- Exactly what the developer needs to do right now
- What happens next
Never explain the debugging process. Never narrate your reasoning. Just direct the next action.
STOP marker: Wherever you see STOP — send your response and wait for developer input before continuing.
Mode picker (front door for empty invocations)
Front door A — empty input → interactive picker
Trigger: $ARGUMENTS is empty.
Use the AskUserQuestion tool. Native UI is faster than reading the help file and printing a menu.
question: "How do you want to describe the bug?"
header: "Input type"
multiSelect: false
options:
- label: "Text description only"
description: "Plain prose only. Example: 'login button does nothing after submit' — tool analyzes the code path, adds trace logs at strategic layers, then you reproduce."
- label: "Text + screenshot"
description: "Visual evidence + your description. Example: screenshot:/tmp/broken-layout.png with 'profile header overlaps the back button on iPhone X'. Tool reads the image as evidence."
- label: "Text + log file"
description: "Console/logcat dump + description. Example: logs:/tmp/logcat.txt with 'app crashes on launch after Firebase init'. Tool parses TRACE_* and ERROR entries."
- label: "Text + screenshot + log file"
description: "Everything from a single reproduction. Example: visual bug + logs captured at the moment it happened. Strongest signal for the tool — use when you have both."
After the answer, prompt for the actual inputs as plain text:
| Choice | Follow-up prompt |
|---|---|
| Text only | "Describe the bug:" |
| Text + screenshot | "Describe the bug, then paste the screenshot path on the next line (or include screenshot:/path inline):" |
| Text + log file | "Describe the bug, then paste the log path (or include logs:/path inline):" |
| Text + both | "Describe the bug, then attach both as screenshot:/path and logs:/path:" |
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 · 404 lines · 15 tokens per session scan A 46055498e045
trace is a command published in the GitHub repository shiyas1331/devkit (4 stars, last pushed 1mo ago), licensed MIT. It adds 15 tokens to every session and 3,926 once invoked, about $0.0001 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 commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.