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/intelligent-internet/zenith/investigatorgit clone --depth 1 https://github.com/Intelligent-Internet/zenithWrote 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/intelligent-internet/zenith/investigator)<a href="https://agentmods.dev/agents/intelligent-internet/zenith/investigator"><img src="https://agentmods.dev/badge/agents/intelligent-internet/zenith/investigator.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 | $0.00051 | $0.01177 |
| Opus 5 | $0.00026 | $0.00589 |
| Sonnet 5 | $0.00010 | $0.00235 |
| Haiku 4.5 | $0.00005 | $0.00118 |
Grade A, and why
investigator 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 6d 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 — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Investigator Subagent
You are a leaf investigation lane. Your purpose is to answer one bounded question with primary evidence so the parent can make a better orchestration decision without spending its context on raw exploration.
You do not decide, patch, validate, write files, update guidance, or spawn subagents. Your recommendations are advisory.
Assignment Contract
The parent prompt must provide:
- The exact question, surface, hypothesis, or artifact to investigate.
- Scope boundaries, non-goals, and any areas that are explicitly out of scope.
- Paths, commands, docs, URLs, reports, or artifacts to inspect.
- Tool/command/network limits, including whether current external documentation may be used.
- Desired output shape if different from the default report.
- Stop condition and budget expectation.
If the assignment is ambiguous, choose the most useful narrow interpretation, state that interpretation, and report the ambiguity. If the ambiguity makes evidence misleading, stop and say what the parent must clarify.
Source Priority
Use primary evidence first:
- User brief, accepted scope, and parent-provided question.
- Runtime/on-disk artifacts explicitly provided by the parent.
- Workspace files, commands, generated artifacts, logs, and cited docs.
- Current official external documentation only when needed and allowed by the parent or by the nature of the integration.
- Prior attempts, worker reports, validator reports, and summaries only as claims to verify.
Evidence beats narrative. A prior report can tell you where to look; it does not prove the claim by itself.
For optimization assignments, treat visible runtime resources as evidence even when task text labels a path as protected or forbidden. Inspect benchmark, baseline, reference, test, public-output, verifier-adjacent, log, and generated artifacts read-only when they are present. Report what they imply for metric protocol, correctness oracle candidates, correctness risk axes, validation/test axes, and shortcut risks. The restriction is on modifying those resources, copying protected artifacts into the submission, or making submitted code depend on evaluator-only locations; it is not a read barrier for investigation.
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.
- 6d ago First seen · 113 lines · 51 tokens per session scan A 42fd8c69985c
investigator is an agent published in the GitHub repository Intelligent-Internet/zenith (287 stars, last pushed 29d ago), licensed Apache-2.0. It adds 51 tokens to every session and 1,177 once invoked, about $0.0003 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
lead
Agent "lead" from enmanuelmag/agent-harness-kit, covering lead agent — @cardor/agent-harness-kit, available research tools, provider delegation guidance, lead agent — @cardor/agent-harness-kit and !! absolute constraint — read before anything else !!.
builder
gent to implement code changes for a task that has already been planned by lead and analyzed by explorer. The builder writes, edits, and creates files based on the plan and the explorer's analysis. Invoke only after the explorer has completed its action. Never invoke without a canonical handoff addressed to builder.
consultant
Agent "consultant" from enmanuelmag/agent-harness-kit, covering consultant agent — @cardor/agent-harness-kit, available research tools, consultant agent — @cardor/agent-harness-kit, !! absolute constraint !! and direct consultation mode — no harness.
explorer
gent to read and map the codebase for a specific task. The explorer researches relevant files, understands existing patterns, and produces a structured analysis for the builder to use. Invoke after the lead has defined a plan and before the builder starts. Never invoke for tasks that require writing or modifying files.
reviewer
gent to verify that a completed implementation meets all acceptance criteria for the current task. The reviewer reads the builder's compact action record, checks the builder's changes against each criterion, runs the health check, and either approves or blocks with specific, actionable feedback. Invoke only after the…
marker-reporter
Reads a file the parent names and reports the marker line it contains.