mifunedev
Organization🏗️ AGRO — Agent Governance Runtime Orchestrator. A portable home for autonomous coding agents.
Categories
Indexed Skills (39)
blog
Turn a source scenario into a publishable blog post. First supported playbook: Loom/demo-to-blog, where a raw notes file plus Loom screenshots become a polished site post after Advisor briefing, photo/text alignment audit, site convention audit, narrative/fact-check audit, source-faithful image handling (exact URLs or local files downloaded from them), optional social/channel-promotion artifact generation, and build verification. Manual-invoke only because it writes content and may spawn delegates. TRIGGER when: /blog <scenario> invoked; asked to create a blog post from a Loom/video/demo/transcript/screenshots; asked to turn .claude/specs/<name>/ demo.md into an openharness-web blog post.
fanout
Ship a set of related issues as parallel PRs — one isolated worktree and one briefed agent session per unit of work, then merge, verify closure, and tear down. Runner-agnostic: herdr, tmux, or background shells. TRIGGER when: asked to work several issues at once, "ship these issues in parallel", "spawn an agent per issue", "fan out", or clearing a queue before a larger refactor.
prompt-miner
Rank past prompts by session outcome and mine the markers that produce the best sessions. Runs the deterministic mine-traces.mjs engine over Claude + Pi JSONL traces, scores each session by a friction + ground-truth outcome proxy, ranks the initiating prompts, then synthesizes falsifiable prompt markers STRATIFIED by session type and proposes harness memory/identity improvements behind a /retro-style propose-then-confirm gate. A cross-session, data-driven cousin of /retro. TRIGGER when: /prompt-miner invoked, or asked to "mine prompts", "rank prompts by outcome", "what prompt patterns work best", "mine session traces", "find good prompt markers", "analyze prompt quality".
rlm
Context-as-environment decomposition (Layer B of the RLM integration). Treat a large artifact (file / dir / log) as a REPL/filesystem the root agent greps and slices instead of ingesting — partition it into addressable chunks via query-context.mjs --map, RECURSE sub-agent calls over the relevant chunks under a bounded depth/children/step budget, AGGREGATE (piping competing per-chunk answers through /weigh), then PERSIST the recursion trace. The anti-context-rot move: ADDRESS context, don't dump the whole artifact into the window. Reuses .oh/scripts/ralph.sh (the recursion loop) and .oh/worktrees/ (isolated branches) BY REFERENCE — never edits either. Manual-invoke (spawns agents, burns tokens). TRIGGER when: /rlm invoked, or asked to "answer a question over a huge file/log", "decompose a large artifact", "recurse over chunks", "address context instead of ingesting it", "beat context rot on a long input", "RLM <file> <query>".
t3
Start, inspect, or stop T3 Code (`npx t3`) in the Open Harness sandbox. Use this for the browser-based T3 Code harness on port 3773, including tmux launch, pairing URL discovery, logs, status, and shutdown. T3 Code wraps an already-authenticated Claude Code, Codex, or OpenCode backend. TRIGGER when: user asks to run T3 Code, start `npx t3`, open the T3 browser UI, get the T3 pairing URL, check T3 Code status/logs, or stop T3 Code.
weigh
Weighted-trajectory selection: sample N candidate trajectories for one task, attach deterministic signals to each, then weight + select among them with a pure, version-controlled scorer we own (score-trajectories.mjs) rather than a model black box. The Workflow tool's judge-panel / adversarial-verify patterns PROPOSE the trajectories; this skill owns the weight function that PICKS among them (frozen DEFAULT_WEIGHTS: consistency:30 evalPass:20 auditPass:15 cost:10 judge:25, sum 100; judge:0 → fully deterministic). disable-model-invocation because the sampling step spawns N agents and burns tokens — manual-invoke only. TRIGGER when: /weigh invoked, or asked to "sample N approaches and pick the best", "weigh these trajectories", "best-of-n this task", "score candidate outputs", "self-consistency vote over samples", "ensemble these answers".
agent-browser
Open a URL in the headless agent-browser, with a preflight health check. Verifies agent-browser is installed and Chromium launches before navigating. Errors out with a diagnostic trace if anything fails. TRIGGER when: asked to open a page, browse a URL, take a screenshot, or test a site with agent-browser.
approve
The approve (approve|deny) decision gate of the canonical workflow — read a spec's critique findings and decide go/no-go BEFORE any GitHub-side state exists, then emit a single APPROVED/DENIED verdict. Auto-gates on critic SEVERITY (any unmitigated high → deny), with an optional human override for borderline findings. This is the decision half of the critique→approve pair; the critics that produce the findings are the separate `/critique` node. TRIGGER when: a plan/PRD has been critiqued and needs a commitment decision before implementation; "approve this spec", "should we build <slug>", "gate <slug> on its critique".
audit
Explicit nine-target audit dispatcher for implementation promotability, one PR, the open PR queue, harness health, context budget, skill integrity, eval quality, drift, and correlated full campaigns. TRIGGER when: audit this task; verify this implementation; audit PR N; classify this pull request; audit open PRs; triage the PR queue; audit the harness; find harness improvements; audit context budget; ablate this context file; audit skills; find stale or broken skills; lint evals; find Goodharted probes; check framework drift; cron staleness; run a full audit campaign; audit everything; cross-target next actions.
autopilot
Self-improvement loop: each hourly run implements the oldest open issue labeled `autopilot` that has no open PR. When the queue is empty it runs first-principles `/audit harness` research, files its own `autopilot` ticket from the top-ranked finding, and builds that. Decomposes via the pm agent, then runs `/ship-spec --issue`, which owns the build end-to-end (the two compacts bracketing implement, a worktree Advisor, `/delegate` workers running `.oh/scripts/ralph.sh`, the `/eval` gate, and the `/audit pr` promotable undraft) through to a ready-for-review PR whose description states why this item was selected. Harness-infra only (skills/rules/docs/scripts/crons/wiki) — never sandbox application code. Runs in its own per-run tmux session. Caps: 6 open autopilot PRs created per UTC day AND 10 total open; never auto-merges. TRIGGER when: the hourly .oh/crons/autopilot.md fires, or invoked manually on demand (e.g. /autopilot --dry-run to preview the next selection).
benchmark
Progress-ceiling verdict gate (part of `/spec execute`'s improve tail) — decide whether ONE landed change was actually BENEFICIAL (moved or held the capability-benchmark ceiling without breaking the regression floor, and is worth its complexity), then emit a single BENEFICIAL/NOT-BENEFICIAL verdict. Composes (never forks) the existing instruments: /eval (the regression floor — probes stay green) + the capability-benchmark ceiling delta (.oh/evals/capability/RESULTS.md vs. the counterfactual). Machinery added with no benchmark movement is NOT-BENEFICIAL by definition. Distinct from /audit implementation (per-unit promotability = floor) — this is the ceiling: did the harness get BETTER, not just not-broken. TRIGGER when: a change has landed and the loop needs a benefit-vs-counterfactual verdict before the cycle repeats; the improve tail of `/spec execute` runs; "was <change> beneficial", "score the capability benchmark", "benchmark this cycle".
builder
Author and refine provider-portable agents, reference skills, task-style command skills, and path-scoped rules using one repository-grounded workflow. TRIGGER when: asked to create, build, scaffold, convert, review, or update an agent, skill, command, workflow, rule, coding standard, or contextual instruction.
cloudflared
Start or explain a Cloudflared tunnel for a sandbox app port. Cloudflared is the default public sharing method for Open Harness previews; this skill replaces generic sharing guidance with a portable pointer to the installed cloudflared CLI and tmux process convention. TRIGGER when: asked to share a local app publicly, expose a sandbox port, make localhost reachable from another machine, open a preview URL, or run cloudflared.
critique
The critique node of the canonical workflow — run two adversarial critics in parallel (implementer lens + user lens) against a freshly-planned spec, write their findings to .oh/tasks/<slug>/critique.md, and hand off to the approve gate. This is the EVIDENCE half of the critique→approve pair; it produces findings, it does not decide (the /approve gate decides). Runs on local artifacts only, before any GitHub-side state. TRIGGER when: a PRD/plan exists and needs adversarial review before the commitment gate; "critique this spec", "run the critics on <slug>", "review prd before we build".
delegate
TRIGGER when: asked to delegate work, execute or parallelize a plan, "run this plan", "delegate this", or after /prd or plan creation. Decomposes work by dependency, launches worker sub-agents in parallel waves, validates completion, and reports results while preserving failure isolation and recursion limits.
firstmate
The `firstmate` build executor — launch ONE long-lived First-Mate session over a whole `.oh/tasks/<slug>/` task graph through the herdr → tmux → foreground runner ladder, where ralph launches 50 fresh single-story processes. Opt-in only, via `--executor=firstmate`; `ralph` stays the default. Documents the executor contract, the ladder and its detection gates, the naming contract, the session budget, the watch matrix, the recovery matrix, and the manual kill procedure for all three runner modes. TRIGGER when: launching or watching an `--executor=firstmate` build, "run firstmate for <slug>", a wedged or stuck firstmate session needs killing, a `FIRSTMATE-INCOMPLETE` line needs interpreting, a firstmate lock needs clearing, or the herdr/tmux/foreground degrade behavior needs explaining.
git
Open Harness git workflow: issues, branches, commits, PR titles/bodies, changelog discipline, worktrees, branch catch-up, stacked PRs, releases, and post-push CI checks. TRIGGER when: any chat mentions git, GitHub, branches, commits, pushes, pulls, PRs, issues, worktrees, merge conflicts, dirty workspaces, changelog entries, release branch/tag workflow, or project git conventions.
health-check
Triage memory, swap, disk and CPU where you are running, and rank Docker reclaim levers by safety×yield. Docker triage is host-only: inside a sandbox with no Docker socket the skill states that once and emits a procedure for the orchestrator to run at the host project root, rather than failing per command. TRIGGER when: asked for a health check, "do we have enough memory/disk" (container-scope inside a sandbox), before starting a heavy stack or docker build (emits the host procedure when the socket is absent), "system health", or "free up space" / "reclaim resources" — which hand you the host block rather than reclaiming inside the container.
herdr
Drive the Herdr terminal workspace manager from the CLI inside the Open Harness sandbox — inspect and control workspaces, tabs, panes, agents, and git worktrees over the Herdr socket API, and read or steer other running agents headlessly. TRIGGER when: asked to run a `herdr` command, list/read/send to another agent or pane, check what agents are running or whether one is idle, split/focus/close panes or tabs, open or create a worktree in Herdr, wait for a pane's output or an agent's status, check Herdr server/session status, or install a Herdr provider integration. Do NOT trigger for tmux-managed headless services (cron, Slack gateway, watchdog) — those stay in their own tmux sessions.
imagine
One-shot draft PRD sketch from a fuzzy scenario. Writes a single artifact to .claude/specs/<slug>/spec.md — a path that is gitignored by default (`.gitignore:51`), so the spec never enters git history. Includes a mermaid diagram. Output is purpose-built as input for `/ship-spec --plan .claude/specs/<slug>/spec.md`, which bypasses /prd's clarifying questions when a plan is provided. No clarifying questions in /imagine itself — generate directly from the <scenario> argument. TRIGGER when: /imagine invoked, or asked to "sketch a spec", "imagine a feature", "draft a quick PRD", "what if we built X".
interview
Adaptive pre-work clarifier. Picks 2–4 task-specific questions via AskUserQuestion, echoes a brief scope summary, then proceeds. Refuses for trivial tasks. TRIGGER when: /interview invoked explicitly, OR before non-trivial work where scope/intent/constraints are ambiguous AND the user hasn't already specified them.
release
Release a validated Open Harness commit by pushing it to main or master, then monitor the automatic SemVer/GHCR/GitHub Release workflow. TRIGGER when: asked to release, version, ship, cut a release, or verify release artifacts.
render-html
Render an artifact (or in-context material) as a bespoke, self-contained HTML file for one-shot human consumption. Writes to .oh/memory/<UTC-date>/<slug>.html. Output is gitignored — these are consumption artifacts, not source. TRIGGER when: asked to render HTML, generate an HTML report, visualize an audit/council/lint/digest, "make this readable", "make a dashboard for", or as a follow-up to /audit harness, /strategic-proposal, /audit skills.
retro
Scientific session-closing retrospective: scan the current conversation, turn each signal into a falsifiable hypothesis, cite session evidence for AND against it, assign a verdict (supported / refuted / inconclusive) and a confidence level, then promote only supported, sufficiently-confident hypotheses into the harness memory tiers (.oh/memory/MEMORY.md, .oh/context/IDENTITY.md) behind a propose-then-confirm gate. Reflects on six learning/knowledge subsystems through the lens of this session — continual learning, context compression, reinforcement learning, wiki, docs, and memory scaffolding — and points at the deep-dive lint/audit skills rather than running them. Operationalizes the Memory Improvement Protocol (.oh/skills/retro/references/memory-protocol.md) as an explicit, evidence-driven, session-closing skill rather than a per-run afterthought. Always appends a log entry. TRIGGER when: /retro invoked, or session closing with decisions, surprises, or failures worth preserving.
ship-spec
Compose /prd → wiki alignment + DeepWiki comparison → 2 critics → /ralph → gh issue → branch → draft PR → implementation/eval/wiki revision/CI → ready-for-review PR into a single end-to-end invocation. Produces a fully-scaffolded task, opens a draft PR for observability while work is pending, then marks the PR ready after the implementation, eval, wiki, and CI gates pass. TRIGGER when: asked to scaffold a spec end-to-end, "ship a spec", "set up a task PR", or after planning a feature and ready to formalize.
spec
Dispatcher for the canonical decomposed workflow (AGENTS.md § The Workflow) — routes the first token of $ARGUMENTS to one of four subcommands: plan, critique, execute, or retro. Each is pointed at a .oh/tasks/<slug>/ folder (the universal interface) and is the same independently-runnable, fan-out-able node /ship-spec decomposes into. Full per-subcommand procedures live in references/{plan,critique,execute,retro}.md. Authority: AGENTS.md § The Workflow. TRIGGER when: a topic/plan/issue needs to become a buildable task folder, "plan <topic>", "scaffold the task for <issue>" -> plan; a planned .oh/tasks/<slug>/ folder needs adversarial review + a go/no-go decision, "critique <slug>", "vet the plan for <slug>" -> critique; an APPROVED .oh/tasks/<slug>/ folder needs building to a promotable PR, "execute <slug>", "build <slug>" -> execute; a build PASSed audit and its lessons should be captured, "retro the <slug> build", "capture lessons for <slug>" -> retro.
ste
Write and rewrite technical prose in Simplified Technical English style: short declarative sentences, active voice, imperative procedures, one term per concept, no hedges, and explicit conditions. Never simplifies code, commands, identifiers, paths, or quoted literals. Marks missing values with a placeholder rather than inventing one. Ships a deterministic checker. TRIGGER when: asked to write, rewrite, simplify, clarify, tighten, or review documentation, a README, a runbook, a spec, release notes, a PR or commit body, an error message, CLI help text, an API description, or agent instructions; or when asked for plain English, controlled language, STE, Simplified Technical English, or "make this unambiguous".
sync
Dispatcher for bidirectional origin↔upstream harness sync — routes the first token of $ARGUMENTS to one of three subcommands: publish (origin→upstream), catchup (upstream→origin), or status (read-only direction check). The canonical topology and intentional divergences to preserve live in references/topology.md. Full per-subcommand procedures live in references/{publish,catchup}.md. This dispatcher composes /audit drift (framework drift detection), /eval (oracle floor), /audit pr (promotability), and /git (branch/PR/CHANGELOG conventions). It NEVER reimplements drift detection — /audit drift owns that. TRIGGER when: "sync to upstream", "publish fork changes to the public repo", "push to mifunedev", "origin→upstream" → publish; "pull from upstream", "port a feature from upstream", "catchup from mifunedev", "upstream→origin" → catchup; "check sync status", "how far behind are we", "which direction needs sync" → status.
teach
Post-implementation communication and learning pass. Reads the completed task artifacts and relevant wiki entry, revises or proposes wiki updates when the implementation changed the provisional model, then teaches the operator the final mental model with verification evidence and understanding checks. TRIGGER when: /teach invoked, "teach me this change", post-implementation understanding, operator communication, or after implementation/audit when the maintainer needs the final model rather than another build step.
watchdog
Generic watchdog for stuck or stale automation. Inspects one or more action classes, determines whether they are merely slow vs actually stuck/stale, and takes the smallest safe completion action. Includes the autopilot draft-PR watchdog: stale draft PRs should be completed and marked ready for review, not merely surfaced as merge nudges.
wiki
Dispatcher for the harness wiki knowledge base — routes the first token of $ARGUMENTS to one of three subcommands: ingest, query, or lint. The corpus lives at .oh/skills/wiki/corpus/ (entity pages) and corpus/raw/ (immutable snapshots), owned by this skill and gitignored-by-default (curated entries are whitelisted into git with `git add -f`). Canonical schema: .oh/skills/wiki/references/schema.md. Full per-subcommand procedures live in references/{ingest,query,lint}.md. Always logs per the Memory Improvement Protocol. TRIGGER when: "add to wiki", "capture this page", "snapshot this source", "ingest <url|path>", or promoting a sub-agent draft -> ingest; "what does the wiki say about X", "find wiki entries for X", "look up X in the wiki" -> query; "lint the wiki", "regenerate the wiki index", "find stale/orphaned wiki entries" -> lint.
worktrees
Manage .oh/worktrees/ lifecycle: create worktree, list worktrees, remove worktree, clean worktrees, stale worktrees audit, isolate work, project clone. TRIGGER when: any git worktree operation, branch isolation needed, stale worktrees review, project clone under .oh/worktrees/project/ (e.g. "clone <owner>/<repo> to worktrees", "add <repo> to .oh/worktrees", "clone this repo into worktrees"), worktree cleanup. A leading-slash harness dir like "/worktrees" still means the repo-relative .oh/worktrees/ — never a literal filesystem-root path.
ci-status
Check the CI pipeline status for the current branch after pushing changes. Reports pass/fail with failure details. Use this after every push to confirm your changes are truly done — CI must be green. TRIGGER when: after git push, after committing changes, when asked to check CI, or when verifying that work is complete.
eval
Run the context fitness-function probe suite (.oh/evals/probes/*.sh) against real state and write the .oh/evals/RESULTS.md benchmark. Each probe is a deterministic 3-state oracle (PASS/REGRESSION/SKIPPED); a green→red transition is surfaced as a REGRESSION naming the lesson it closes. Tier-B behavioral evals are out of scope. TRIGGER when: asked to run evals, check the probe suite, "run /eval", verify a lesson's probe is green, benchmark the harness, or before/after editing a rule/skill that a probe guards.
harness-context
Explain the Open Harness architecture, layout, and conventions. Use this skill when the user asks about project structure, layout, git workflow, rules, or how things are organized. Returns specific file paths and `path:line` citations. TRIGGER when: asked about harness structure, layout, conventions, rules, process, or project organization.
post-bridge
Interact with the Post Bridge social media post management API. Use when the user wants to publish posts, upload media, manage social accounts, schedule content, or retrieve analytics via the Post Bridge platform. Triggers on: post to instagram, upload media to post bridge, schedule a social post, list social accounts, create a post, publish video, post bridge api.
prd
Generate a Product Requirements Document (PRD) for a new feature. Use when planning a feature, starting a new project, or when asked to create a PRD. Triggers on: create a prd, write prd for, plan this feature, requirements for, spec out.
ralph
Convert PRDs to prd.json format for the Ralph autonomous agent system. Use when you have an existing PRD and need to convert it to Ralph's JSON format. Triggers on: convert this prd, turn this into ralph format, create prd.json from this, ralph json.
strategic-proposal
Spawn 5 domain experts to propose roadmap items, then an AI council drafts a roadmap, a Strategic Critic challenges it, and the council finalizes with revisions. Updates the pinned roadmap issue and /roadmap page data. TRIGGER when: asked to build roadmap, prioritize features, strategic proposal, "what should we build next", or rank product priorities.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.