Borrowing it
Nothing to install: this file belongs to Traves-Theberge/Orchestra. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Traves-Theberge/Orchestra/main/.codex/skills/maestro-integrations/SKILL.mdgit clone --depth 1 https://github.com/Traves-Theberge/OrchestraWrote 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/traves-theberge/orchestra/maestro-integrations)<a href="https://agentmods.dev/skills/traves-theberge/orchestra/maestro-integrations"><img src="https://agentmods.dev/badge/skills/traves-theberge/orchestra/maestro-integrations/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/traves-theberge/orchestra/maestro-integrations"><img src="https://agentmods.dev/badge/skills/traves-theberge/orchestra/maestro-integrations.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00037 | $0.00800 |
| Opus 5.5 | $0.00015 | $0.00320 |
| Sonnet 5.5 | $0.00007 | $0.00160 |
| Haiku 4.5 | $0.00004 | $0.00080 |
Grade A, and why
maestro-integrations 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 19d 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 — 30 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Maestro integration journeys
Use this skill for cross-project issue discovery, assignment, task execution, and PR linkage. Orchestra is the application; Maestro is the agent.
Resolve the selected backend and exact registered project using orchestra-cli or orchestra_control. Read the project's issue source and tracker connection before choosing a provider. A Git remote or GitHub owner/repository does not make local SQLite tasks into GitHub issues. Blank issue source with no tracker connection currently means SQLite. Copy returned project/task IDs in full into commands; display abbreviations are not executable identities.
Read the matching provider skill under .agents/skills in Maestro's control profile. Maintained repository copies live under .codex/skills; use the control-profile copy during a Maestro session:
maestro-github/SKILL.md: repository issues and GitHub PRs, including GitHub PRs attached to Linear/Jira tasks.maestro-linear/SKILL.md: Linear team/issue identity and workflow state.maestro-jira/SKILL.md: Jira site/project/issue identity and workflow mapping.maestro-azure-devops/SKILL.md: Azure Boards/Repos identity and the currently missing adapter.
Keep the journey's identities together: backend, Orchestra project ID, source/connection, external issue ID and URL, task ID, workspace ID/path, requested worker/harness, run observations, and exact PR URL. An issue URL is not a PR URL; a worktree or conversation is not the task. Preserve observed pr_url through task reads and assignment. Report absent linkage rather than constructing a URL from an issue number or branch.
For a supported task flow, observe the exact source inventory, create or resolve the Backlog task, configure/assign it using advertised shared controls, then queue it separately when complete. Retain one UUID and inspect its receipt after uncertainty. Queue acceptance is not proof of a running agent. Observe workspace/run state, then PR head/checks/review state independently. A merged PR does not establish a Done task or authorize checkout deletion.
Use the configured backend for issue-backed task mutations so source routing and receipts are retained. Repository/PR operations may use an available authorized provider tool or CLI, with explicit repository identity. Do not bypass an unavailable shared task mutation using a guessed HTTP endpoint. Never copy credentials into agent files, prompts, issues, or skills.
The pipeline contract applies to every harness. Todo plans and waits for explicit human approval of the returned exact plan hash. Maestro may forward a human's explicit approval through an advertised shared control, but must never infer approval from planning completion, agent feedback, or a request to fix the pipeline. In Progress executes only the approved current context. Replanning retains the existing code, worktree, PR and feedback while invalidating the old approval. A different provider does not bypass these gates; verify that its adapter supports the required stage controls before dispatching.
PR review is a separate agent stage bound to the exact repository, PR and current head commit. Preserve reviewer identity and feedback. Requested changes return to planning with that context; a clean agent review waits for human approval. Never treat a review result as human approval, merged-head evidence, or permission to mark Done. Use only review controls advertised by the selected executable; missing controls are an implementation gap, not permission to imitate them with generic task state patches.
Connection configuration, adapter presence, and fixture tests are separate from a successful authenticated read. Report authentication failure as connection failure, not an empty inventory. Hosted adapters currently have routing/state/assignment limits; a source skill describes them and does not enable a missing operation. Do not change the selected tracker or fall back to a different provider to make a request appear successful.
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.
- 19d ago First seen · 30 lines · 37 tokens per session scan A f401ad409c4b
maestro-integrations is a skill published in the GitHub repository Traves-Theberge/Orchestra (26 stars, last pushed 10d ago), licensed Apache-2.0. It adds 37 tokens to every session and 800 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-06.
Other skills, from other repositories
agtx-oneshot
One-shot a whole project on an agtx board: decompose the goal, run every task unattended, unblock the workers, and merge each one. Use when the user wants a long autonomous run rather than a single session.
agtx-sweep
Sweep this conversation into agtx tasks and push them to the kanban board. Use when the user wants to capture, decompose, or hand off conversation results to the agtx board.
symphony-skill
Single Symphony operator router for Kanban tickets, service/TUI runs, WORKFLOW.md and prompt config, lane customization, delegation, production-ready app delivery planning, OneShot runs, repo and monorepo bootstrap, and worker failure triage.
kagan
Read a ticket, link threads, and move work through the native Kagan board when the user asks for ticket work.
superpm
Produce focused PM artifacts and verify them with a critic. Use for PRDs, strategy, OKRs, roadmaps, specs, research, analytics, GTM, growth, pricing, positioning, market signals, release notes, retros, and resume reviews.
Mochi-a-pm
Use this skill when the user or another AI needs project management, task tracking, priority assignment, timeline planning, or progress review. Mochi is a cute but deadly serious project manager AI.