Chief of Staff

Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else. Prerequisites: An agent environment that can launch subagents (the skill does all work through subagents, in the background where possible). No account, no API key, no paid service and no network access is required by the skill files themselves. Heads up, disclosed here: this is an upstream in progress skill (folder skills/in-progress/), it is invoked only explicitly (disable-model-invocation: true), a session running it is meant to run for a long time, and it will suggest recurring schedules and proactively improve the working environment (standards files, lint rules, data source access such as logs, test environment databases and, when necessary, browser access) as part of its method: each of those steps still needs your own agent environment's normal approvals. Nothing is sent anywhere by the skill itself. Install "Chief of Staff" for me. A chief of staff for a long running goal pursued inside a single agent session. The skill makes the agent the directly responsible individual for the goal and has it think on two tracks at once: tactical (complete the immediate task) and strategic (modify the environment so the next task goes better: constrained APIs and functions, lint rules, coding standards files, data sources such as logs and test databases, and a no workarounds rule). All work is done in subagents, in the background where possible, communicating sparsely through context pointers (research notes, previous commits) instead of duplicated information, and the skill suggests recurring schedules when the harness permits. Upstream state, stated plainly: the skill lives in the repository folder skills/in-progress/, its frontmatter sets disable-model-invocation to true, and its OpenAI agent file sets allow_implicit_invocation to false, so it runs only when you invoke it explicitly. It is an in progress skill, not a finished one. Repository: https://github.com/mattpocock/skills (the skill is the whole skills/in-progress/chief-of-staff/ folder at the pinned commit: SKILL.md, agents/openai.yaml) Pinned version: install the code at commit 49dd158d1076134a641b33efb035946536778336 exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/mattpocock/skills into a temporary folder, check out commit 49dd158d1076134a641b33efb035946536778336, and copy the skills/in-progress/chief-of-staff/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-18: [FINGERPRINTS] - SKILL.md: SHA-256 6cc6e47ab5022f3e45e5be0fedbb0f6e645f4f0190d6b0c51c7cd60c2d7c1d98 - agents/openai.yaml: SHA-256 c0f68ed9da6977adc97a791be3bb0830d301b2d27f9c635d92d6f325bffefb83 If a file does not match, it changed since the listing was prepared: do not install it yet, tell me what you found and let me decide. 2. Read SKILL.md in full (and every other file in the folder) and tell me in plain words: what the skill does when used, which services it connects to, whether it reads or sends any personal information or keys, and anything that seems unusual. Stop on any red flag and tell me before going further. Confirm these points from the files: - The skill folder holds 2 files: SKILL.md and agents/openai.yaml (display name and policy only). There is no executable code, no script and no tool of its own. - Nothing is sent anywhere during installation itself; the only download is the skill folder from GitHub at the pinned commit. - The frontmatter sets disable-model-invocation: true and openai.yaml sets allow_implicit_invocation: false: the skill starts only when you invoke it explicitly, never on its own. - At use time the work happens inside your own agent environment through subagents you can see: the method suggests schedules and environment improvements (standards files, lint rules, logs, test databases, browser access when necessary), it does not grant itself any of those accesses. 3. When I say "yes, install", install it as a skill: copy the whole folder into the agent's skills directory, in a folder named "chief-of-staff", keeping every file (SKILL.md, agents/openai.yaml). 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (for example: sign in or create an account, provide any API key, or install any local tool named in the prerequisites; using the skill is a separate step that only starts when I ask for it, and every paid or authenticated use disclosed above happens only then). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.