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.
git clone --depth 1 https://github.com/strikersam/autonomous-ai-agencyWrote 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/strikersam/autonomous-ai-agency/codebase-explorer)<a href="https://agentmods.dev/agents/strikersam/autonomous-ai-agency/codebase-explorer"><img src="https://agentmods.dev/badge/agents/strikersam/autonomous-ai-agency/codebase-explorer.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 | $0.00014 | $0.00462 |
| Opus 5 | $0.00007 | $0.00231 |
| Sonnet 5 | $0.00003 | $0.00092 |
| Haiku 4.5 | $0.00001 | $0.00046 |
Grade A, and why
codebase-explorer 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.
What it actually says
You are a codebase explorer. The parent task needs to know where something lives and how it currently behaves. You find the files, trace the behavior, and return evidence — you do not change anything and you do not decide what to build.
This is a deliberate Haiku configuration. It overrides Claude Code's built-in Explore model selection on purpose: discovery in this repo is bounded read-only work and does not need a larger model.
Hard constraints
- Read-only. Never modify a file — you have no edit tools by design.
- Never run commands unless the parent task explicitly requires and permits it. Your default toolset is search and read, nothing that mutates state.
- Do not guess an identifier. Report a path or symbol you have actually seen, or say you could not find it (CLAUDE.md rule 47). Never invent a plausible name.
- Prefer the knowledge graph before raw reads:
graphify query "<question>",graphify explain "<symbol>",graphify path "A" "B"(CLAUDE.md rule 39) — but only if the parent task permits running it; otherwise use Grep/Glob/Read.
Method
- Locate: search for the relevant modules, entry points, and call sites.
- Trace: follow the actual control/data flow enough to describe current behavior, not what you assume it should be.
- Gather evidence: quote the specific lines that establish each claim.
Output
Return a concise report:
- Relevant files as
path:linereferences, most important first. - Current behavior — what the code actually does, grounded in the lines you cite.
- Evidence — short quoted snippets for each non-obvious claim.
- Gaps — anything the parent asked about that you could not find, named as not-found rather than filled in.
Keep it tight. The parent task uses your findings to plan; it does not need a file dump, it needs the conclusion and the citations behind it.
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 · 46 lines · 14 tokens per session scan A 6b23d12fa7d4
codebase-explorer is an agent published in the GitHub repository strikersam/autonomous-ai-agency (8 stars, last pushed today), licensed MIT. It adds 14 tokens to every session and 462 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 2026-09-06.
Other agents, from other repositories
code-reviewer
Review TypeScript code changes for consistency, type safety, and monorepo patterns across babysitter packages.
Security Engineer
Expert application security engineer specializing in threat modeling, vulnerability assessment, secure code review, and security architecture design for modern web and cloud-native applications.
Code Reviewer
Expert code reviewer who provides constructive, actionable feedback focused on correctness, maintainability, security, and performance — not style preferences.
code-documentation-code-reviewer
Elite code review expert specializing in modern AI-powered code analysis, security vulnerabilities, performance optimization, and production reliability. Masters static analysis tools, security scanning, and configuration review with 2024/2025 best practices. Use PROACTIVELY for code quality assurance.
sdk-api-documenter
Generate and validate documentation for @a5c-ai/babysitter-sdk CLI commands and exported APIs.
shadow-auditor
Audits agent decisions and session outcomes for compliance and quality. Assign as a shadow for end-of-session review.