Bootstrap
Initialize explicitly requested, missing AgentOps documentation and optional verdict storage without taking over repository workflow. Triggers: "bootstrap AgentOps", "initialize AgentOps docs".From its SKILL.md
npx -y skills add boshu2/agentops --skill bootstrapAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
SKILL.md
3.3 KB, as published. Nobody here has run it
Bootstrap — minimal project setup
Bootstrap fills only explicitly requested, missing AgentOps entry documents and, when requested, the durable verdict directory. It does not initialize Git, install hooks, create tracker state, start runtimes, or impose a delivery workflow.
Never-overwrite is what makes bootstrap safe to run on any repository: a setup step that can only add is idempotent by construction, while one that can replace must first prove it understands what it is replacing.
Named failure mode — scaffold sprawl: creating files the caller never requested because a "complete" setup feels more helpful than a minimal one.
Anti-pattern: inferring product intent from directory names and READMEs to avoid asking the caller. Corrective: ask for the missing content; a wrong PRODUCT.md written confidently is worse than a question.
Procedure
- Inspect the target directory and report which canonical files already exist.
- Ask the caller for missing product intent or goal content when it cannot be inferred safely.
- Create only missing, explicitly requested files. Never overwrite an existing document.
- Create
.agents/ao/verdicts/sha256/when durable local verdict storage is requested. - Validate filesystem existence and report created, skipped, and failed paths.
- Stop.
Typical documents are PRODUCT.md, GOALS.md, AGENTS.md, and a README section
that explains the RPI traversal. Generated product copy starts from the
operations-layer category and preserves the ownership boundary. Repositories
remain free to use their own Git, CI, tracker, release, and deployment
policies.
Naming. Three surfaces share the word "bootstrap"; they are distinct. This
skill authors missing entry documents. ao init is the CLI command that creates
the local evidence and verdict directories (.agents/ao/**). ao session bootstrap is a read-only session command that reports which local orientation
files are present. This skill invokes neither.
Non-goals
- installing or invoking
ao,br,bd, NTM, Agent Mail, or another runtime; - creating
.git, worktrees, branches, commits, hooks, or CI workflows; - choosing work or claiming that repository setup is complete beyond the paths actually inspected;
- running RPI automatically.
Output
Return target path, requested files, created files, existing files left intact, failed writes, and validation observations. Do not include a next action.
References
What ships with it: 1 file
1.2 KB alongside SKILL.md
references/
- examples.md1.2 KB
Gives 0 of the 12 instructions most docs writing skills give
Counted across 1,951 of the 3,904 authors here whose files we hold, read 2026-09-06
- Use third-person for skill descriptionsin 54 of 1951, across 35 files
- Start descriptions with Use whenin 43 of 1951, across 29 files
- Run baseline scenarios before writing any skillin 40 of 1951, across 26 files
- Use active voicein 40 of 1951, across 36 files
- Map file responsibilities before defining tasksin 36 of 1951, across 29 files
- Use checkbox syntax for tracking stepsin 35 of 1951, across 27 files
- Ask one question at a timein 35 of 1951
- Offer execution options after saving the planin 33 of 1951, across 24 files
- Include complete code in every stepin 33 of 1951, across 27 files
- Design units with clear boundaries and interfacesin 31 of 1951, across 23 files
- Announce the skill usage at the startin 30 of 1951
- Verify agent compliance after adding the skillin 29 of 1951, across 17 files
Said here and by no other author read
- Inspect target directory for existing canonical files
- Ask caller for missing product intent or goals
- Create only missing explicitly requested files
- Create verdict directory if requested
- Report created skipped and failed paths
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once.