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 skills add ReinaMacCredy/maestro --skill maestro-exploregit clone --depth 1 https://github.com/ReinaMacCredy/maestroWrote 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/skills/reinamaccredy/maestro/maestro-explore)<a href="https://agentmods.dev/skills/reinamaccredy/maestro/maestro-explore"><img src="https://agentmods.dev/badge/skills/reinamaccredy/maestro/maestro-explore/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/reinamaccredy/maestro/maestro-explore"><img src="https://agentmods.dev/badge/skills/reinamaccredy/maestro/maestro-explore.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00079 | $0.00667 |
| Opus 5 | $0.00039 | $0.00333 |
| Sonnet 5 | $0.00016 | $0.00133 |
| Haiku 4.5 | $0.00008 | $0.00067 |
Grade A, and why
maestro-explore 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 5d 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 — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
maestro-explore
Answer an open question with evidence, without touching production code. Read-only toward production paths; exploration never authorizes implementation. Any tier may use it.
Three modes; pick whichever settles the question:
- research - documented or external facts: read source, docs, or the web; record findings with links so claims stay traceable. Procedure: references/research.md.
- prototype - an environment-specific behavior question that reading
cannot settle: build the smallest throwaway that answers it, outside
production paths. Prototypes are disposable; only a decision-encoding
fragment (a schema, a reducer, a type shape) may be inlined into a locked
decision or the bundle's SPEC.md. Fixing the prototype's own bugs is still
this mode - a fix routes to
maestro-workonly when the target is production code; iterate directly, no re-routing pass per request. When the user approves porting, the port is production work: route it by tier (maestro-bundletier rule); the prototype never merges as-is. "Ship it as-is" waives the rewrite, not the route. A prototype that is deployed or relied on in real work is no longer throwaway: say so, and route its next change by tier. Procedure: references/prototype.md. - baseline - current repository behavior that must be preserved or
changed: capture it as runnable commands with observed output, so
maestro-verifycan compare later. Phrase each behavior the work must preserve as a guard, "the system SHALL CONTINUE TO ", so preserved behavior is a checkable claim, not an assumption.
Where findings land
- A fact that settles a fork becomes a decision:
maestro decision draft "<choice>" --rationale "<why, with source link>" --work <id>, thenmaestro decision lock <id>. - Working evidence for an open work item:
maestro work note <id> "<finding + link>"; in a bundle, also the NOTES.md Current State. - A baseline the work must preserve: a VERIFY.md scenario in the bundle, or the work item's acceptance when there is no bundle.
- No work item and no bundle: deliver the findings in the conversation. Open a work item only if the question feeds work that will actually happen; a finding alone never opens a bundle.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 5d ago First seen · 52 lines · 79 tokens per session scan A 3f6ea371314c
maestro-explore is a skill published in the GitHub repository ReinaMacCredy/maestro (231 stars, last pushed 4d ago), licensed MIT. It adds 79 tokens to every session and 667 once invoked, about $0.0004 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 2026-09-05.
Other skills, from other repositories
qveris
Discover and call third-party API capabilities through QVeris, using inspect or probe only when their checks are needed, then generate production REST code. Use when task fit, data quality/freshness, provider comparison, fallback, or the user request favors QVeris.
deepseek-harness
Use when building AI agent applications with a plugin-based architecture — Web UI, CLI, Python SDK, Cordis plugin system, multi-model orchestration. DeepSeek Harness (dsh): open-source agent harness by DeepSeek AI where everything is a plugin, powered by Cordis for spatiotemporal composability.
ui5-typescript-conversion
A skill for converting UI5 (SAPUI5/OpenUI5) projects to TypeScript.
skill-creator
Create, edit, improve, or audit AgentSkills. Use when creating a new skill from scratch or when asked to improve, review, audit, tidy up, or clean up an existing skill or SKILL.md file. Also use when editing or restructuring a skill directory (moving files to references/ or scripts/, removing stale content, validating…
sub-agent-creator
Create, edit, convert, or audit ChatLuna sub-agent markdown files. Use when adding a new sub-agent, refining a sub-agent prompt, choosing a sub-agent's goal and output contract, restricting tools, skills, MCP, or computer permissions, pinning or omitting a model, setting maxTurns, placing agents under local…
agentcli
Use this skill to inspect or change ChatLuna agent admin state — skills, sub-agents, tools, MCP servers, MCP tools, or permission rules. The skill edits a working copy of the agent config inside the sandbox; the user must run chatluna.agent.sync to write changes back to the host instance.