fno

Natural-language remote control for the `fno agents` worker mesh. Describe the outcome you want - spawn or hand off work, message or inspect workers, drive a session, or stop it - and this skill resolves and runs the provider- and substrate-specific CLI command. It normalizes arguments, applies confirmation policy, and reports genuine receipts. Use when: 'spawn a worker for ab-XXXX', 'seed a session with this prompt', 'hand off this doc to codex', 'show my running agents', 'drive the reviewer', 'stop the billing worker'.

Multi-perspective feature completeness analysis and planning loop. Use when: 'audit feature', 'what's missing', 'feature completeness', 'gap analysis', 'discover features to build', 'plan all features'.

Self-improvement loop for the toolkit. Passive capture (git post-commit, verifiers, /insights tags) into ~/.fno/corrections.log; a monthly review surfaces patches; the user triages in about 20 minutes. Use when asked to review corrections, triage proposed patches, install or check the schedule, ingest /insights, or audit recurring mistake classes.

Create an implementation plan as a single .md doc that is a contract for the implementer. Take an idea, a node, or cited findings and produce a BDD-aware plan. One .md == one PR == one node. The frontmatter is locked. The body is whatever makes the plan understood. Use when: 'create plan', 'implementation blueprint', 'break this down', 'how should we build'.

Keep prompt cache alive during idle. Prevents 10x cost spike when cache expires. Use when: 'keep cache warm', 'cache keepalive', or auto-activated at session start when project opts in.

Execute a plan. Routes between a lightweight single-session executor (flat, default) and full wave orchestration (waves, alias operator). Use when: 'do this plan', 'execute the plan', 'run the waves'.

Repair a broken state. Routes to the fast one-fix-per-iteration loop with auto-revert (fix, default) or the scientific-method hypothesis loop (investigate). Use when: 'fix all errors', 'make tests pass', 'fix the build', 'debug this', 'investigate this failure'.

Join this session to the fno mesh so other sessions can message you

Daily levers-only backlog grooming pass. Reads the graph, merged PRs, and the triage pile, applies a fixed allowlist of reversible levers, and mails a one-screen report. Use when: 'groom the backlog', 'daily grooming pass', 'clean up the backlog', dispatched by `fno backlog groom`.

Growth-studio launch orchestrator. Drafts a four-role campaign bundle against verified product truth, runs the factual/brand/accessibility evaluators, and holds at a founder approval gate. Refuses outright when growth-studio is not activated. Dispatches no external effect.

Encode-before-exit ritual for an episodic orchestrator: read the track, write the wave plan, encode it into the graph, kick off, abdicate. You are crowned over one scope, you rule it once, the crown expires. Use when: 'crown me on <epic>', 'orchestrate the backlog', 'plan the next wave', 'king for a day on <epic>'.

Record a durable project law from chat in one step.

Message background agent workers and projects from a runner-less surface (phone / Happy app), or hold this session's incoming mail while the operator needs uninterrupted time. One front door over the shipped `fno agents mail` surface: send, reply, unread/list/view/status, ack, drain, and timed DND hold. Runs the genuine CLI and reports real receipts. Use when: 'send tgt-foo a message', 'check my unread', 'hold my mail', 'turn on DND', 'do not interrupt me', or 'I need your time for ten minutes'.

Grading lenses the blueprint judge reads to argue against a finished plan. Not for drafting.

Drive a PR through its lifecycle. Routes to create (open a PR via a routed pr-create worker), check (poll for external review and implement it), or merged (the post-merge ritual). Use when: 'create pr', 'open pr', 'submit pr', 'check pr', 'get review', 'post merge', 'process the merged PR'.

The tenured king: stay active over a territory for days. Crowned once, check in on a schedule, drive with levers, park rather than die. Composes king-for-a-day (the one-wave pass) with a self-injected beat. Use when: 'reign over <scope>', 'stay king over <epic>', 'keep driving this territory'.

Review a diff or a research brief. Routes to the fno-owned inline review lane (default: levels low/medium/high/xhigh/max, --comment, --fix, optional PR/branch/path target), runtime evidence (prove-it), the apply-or-skip cleanup pass (cleanup), a cross-model second opinion (peer), the advisory research-verify panel for a doc deliverable (research), or a self-cert attestation for the config.review.reviewers gate (declare). Use when: 'review this', 'code review', 'is this ready', 'prove it works', 'clean this up', 'get a second opinion', 'review this research brief', 'declare this reviewed'.

Schema-driven setup wizard for footnote settings. Derives its question set from the Pydantic config model (the single source of truth) instead of a hardcoded list; asks only the real per-project decisions and defaults the rest.

Drive any deliverable to its finish line. The umbrella over delivery terminals: 'ship pr' is the PR lifecycle (= today's /pr), 'ship doc' ships a research brief to output_dir and grades it. Use when: 'ship this', 'ship a PR', 'ship the doc', 'ship the brief', 'deliver this'. Not for ongoing areas (budget, community) - those have no finish line; use /target.

Generate and maintain project documentation - architecture docs, how-to guides, API references. Use when: 'document this', 'architecture docs', 'API contract', 'runbook', 'user guide', 'how-to', shipping documentation for a feature.

Run N parallel variations of the same feature for comparison. Use when: exploring design alternatives, comparing architectures, A/B testing implementations, 'give me 3 takes on this'.

Use when: build this feature, get it done end-to-end, or execute a plan from idea to PR.

Test-Driven Development: write the test first, watch it fail, implement minimal code. Use when implementing a feature or bugfix, during /execute, or when a plan carries acceptance criteria.

Investigate a question against primary sources - repo source code, specs, first-party APIs - and write cited findings to one Markdown file. Use when: 'research this', 'think through this', 'how does this break' (what-if), 'several lenses on this' (panel), 'what owns this' (class). Prefix `bg` or `subagent` to run it off the main thread.

Propose and apply optimal ordering for pending specs on the graph backlog. LLM proposes, human approves. Use when: 'triage the backlog', 'what should I work on next', 'reorder my specs', 'find duplicates in graph'.

Loaded at every SessionStart so the agent knows the two footnote surfaces (slash-command workflows + fno CLI primitives) exist from turn one. Mirrors the session-start pattern.