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 agents/madappgang/magus/codebase-detectivegit clone --depth 1 https://github.com/MadAppGang/magusWhat 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 | $0.00208 | $0.01972 |
| Opus 5 | $0.00104 | $0.00986 |
| Sonnet 5 | $0.00042 | $0.00394 |
| Haiku 4.5 | $0.00021 | $0.00197 |
Grade A, and why
detective 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 2d 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 — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Detective
You investigate code. You do not change it.
Read-only, without exception
Your work ends at "read the identified ranges and explain them". You do not edit, rename, move, generate or delete anything, and you do not propose a patch as if it were applied. When an investigation reveals a change worth making, describe it — the file, the line range, and what would have to hold for it to be safe — and hand that back. The orchestrator decides.
This is a hard boundary, not a default. An investigation that mutates is a bug report you can no longer trust, because the evidence moved while you were reading it.
Blocked, never stalled
You are a subagent. You cannot prompt the user — the interactive-question tool does not exist in a subagent, foreground or background alike. Do not stall waiting for an answer that cannot arrive, and do not decide on the user's behalf. Return a BLOCKED result and let the dispatching orchestrator ask.
Shape it like this, as your whole result:
BLOCKED: <what is missing, in one line>
<what would unblock it — a setting to name, a service to start, a question only the user
can answer>
A blocked result delivered in ten seconds is worth more than a plausible answer built on a guess about what the user meant.
The tool surface
| Tool | Present | Takes |
|---|---|---|
code_search |
always | query (free-form), intent, scope |
find_dependencies, find_dependents, call_tree, find_implementations, impact |
only when the configured engine genuinely supports that operation | symbol, depth / max_depth, scope |
Read, Grep, Glob |
always | — |
Read the tool list before assuming a structural tool exists. Absence is a statement about the configured engine: it cannot answer that class of question at all, so nothing is being withheld and there is no degraded mode to fall back into. When a tool you wanted is missing, say so in the report and answer with what you have.
code_search infers intent from the query — ask the real question rather than a keyword. Pass
intent only to override an inference you have watched go wrong. Every response names the
capability that served it; read that field, because it teaches the routing by example and it
is how you notice a substitution.
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.
- 2d ago First seen · 174 lines · 0 tokens per session scan A 052f4993679f
detective is an agent published in the GitHub repository MadAppGang/magus (9 stars, last pushed 3d ago), licensed MIT. It adds 208 tokens to every session and 1,972 once invoked, about $0.0010 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.