From citadel
Routes user intents to optimal skills, orchestrators, or commands via classification of scope, complexity, persistence, and parallelism. Use /do for unknown tools, status, setup, or skill discovery.
npx claudepluginhub sethgammon/citadel --plugin citadelThis skill uses the workspace's default tool permissions.
Use `/do` when the user wants something done but doesn't know (or care) which tool handles it.
Enforces routing of every agent action through correct skills before code, files, or technical decisions. Checks adaptive overrides from metrics, defaults to rune:cook for code tasks.
Routes natural language requests to skills and workflows by parsing intent and matching patterns for SDLC phases, gate checks, artifacts, security reviews, and marketing tasks.
Indexes skills with 3-layer progressive disclosure: Layer 1 index (~100 tokens) always loaded, Layer 2 summary (~300 tokens) on context need, Layer 3 full skill on execution. Saves 90%+ tokens on discovery.
Share bugs, ideas, or general feedback.
Use /do when the user wants something done but doesn't know (or care) which tool handles it.
Don't use when: you know the destination — invoke /marshal, /archon, /fleet, or any skill directly.
| Command | Behavior |
|---|---|
/do [anything] | Classify intent, route to cheapest capable path |
/do status | Show full harness dashboard (/dashboard) |
/do continue | Resume most recent active campaign or fleet session |
/do --list | Show all skills grouped by category with trigger keywords |
/do setup | First-run experience — configure the harness for this project |
Classification runs top-to-bottom. First match wins. Each tier is cheaper than the next.
Before routing, check if new skills have been added since last registration.
.claude/skills/)registeredSkillCount from .claude/harness.jsonregisteredSkills array from harness.json (default: [])
b. Diff skill names against the registered list
c. For each unknown skill: read ONLY lines 1-10 of its SKILL.md (frontmatter)
d. Extract name and description from frontmatter
e. Add the skill to the Tier 2 keyword table for this session using its
name and description words as match targets
f. Log to the user: "Discovered {N} new skill(s): {names}. Run /do setup to permanently register routing keywords."
g. Update registeredSkillCount and registeredSkills array in harness.jsonThis means:
/do setup does a full registry rebuild with permanent keyword assignmentRegex/keyword on raw input. Catches trivial commands:
| Pattern | Action |
|---|---|
| "typecheck" or "type check" | Run the project's typecheck command |
| "build" | Run the project's build command |
| "test" or "tests" | Run the project's test command |
| "status", "dashboard", "what's happening", "what's going on", "show activity" | Show full harness dashboard (/dashboard) |
| "continue" or "keep going" | Resume active campaign or fleet session |
| "setup" | Run /do setup first-run experience |
| "--list" or "list" | Show all available skills |
| "fix typo in X" or "rename X to Y" | Direct edit (no orchestrator needed) |
| "commit" | Stage and commit changes |
| "rollback", "undo phase", "restore checkpoint" | Find active campaign, read latest checkpoint ref, run git stash pop |
If matched → execute directly. Done.
Check for active campaigns or fleet sessions that match the input scope:
.planning/campaigns/ for files with Status: active or status: active in frontmatter.planning/fleet/ for session files with status: active or needs-continuetype: improve in
frontmatter, route to /improve {target} --continue where {target} is the campaign's
target field. Do NOT route improve campaigns to archon -- improve is its own orchestrator./archon continue/fleet continue.planning/daemon.json exists with status: "running": the daemon spawned
this session but there's no work to do. Update daemon.json:
status: "stopped", stopReason: "no-active-work",
stoppedAt: "{ISO timestamp}". Delete both triggers if IDs are present.
Output: "[daemon] Stopped -- no active campaign found. The work is done."If matched → resume the active work. Done.
Match input against installed skill keywords from Citadel's built-in skills
and any project-level custom skills in .claude/skills/.
Built-in skill triggers:
| Input Contains | Route To |
|---|---|
| "prd", "requirements", "spec", "plan an app", "design an app" | /prd |
| "architect", "architecture", "design the system", "file structure", "plan the build" | /architect |
| "create app", "build app", "build me", "make an app", "new app", "generate app" | /create-app |
| "add [feature]", "implement [feature]", "add auth", "add payments", "integrate [x]" | /create-app (Tier 5 — feature mode) |
| "review", "code review" | /review |
| "test", "generate tests", "write tests" | /test-gen |
| "document", "docs", "docstring", "readme" | /doc-gen |
| "refactor", "rename", "extract", "split file" | /refactor |
| "scaffold", "new module", "new component", "bootstrap" | /scaffold |
| "create skill", "new skill", "repeated pattern" | /create-skill |
| "handoff", "session summary" | /session-handoff |
| "orchestrate", "chain skills", "multi-step" | /marshal |
| "campaign", "multi-session", "phases" | /archon |
| "parallel", "simultaneous", "multiple agents", "at the same time", "both ... and" | /fleet --quick |
| "intake", "process pending", "pipeline" | /autopilot |
| "setup", "first run", "configure harness" | /setup |
| "research", "investigate", "look into", "find out" | /research |
| "experiment", "optimize", "try", "A/B", "measure" | /experiment |
| "debug", "root cause", "diagnose", "why is", "investigate bug" | /systematic-debugging |
| "research fleet", "parallel research", "multi-angle research", "compare options" | /research-fleet |
| "preview", "screenshot", "visual check", "does it render" | /live-preview |
| "postmortem", "retro", "what broke", "what happened", "debrief" | /postmortem |
| "design", "style guide", "design manifest", "visual consistency" | /design |
| "qa", "test the app", "click through", "does it work", "browser test" | /qa |
| "triage", "open issues", "unlabeled issues", "review pr", "review prs", "investigate issue" | /triage |
| "watch pr", "watch ci", "monitor pr", "fix ci", "ci failing", "pr failing", "auto-fix", "auto fix pr", "pr is red", "checks failing" | /pr-watch |
| "dashboard", "what's happening", "what's going on", "show activity", "harness state", "show me status" | /dashboard |
| "telemetry", "what did this cost", "session cost", "how much did that cost", "how much have I spent", "what hooks fired", "trust level", "show me telemetry", "spending", "session stats", "what telemetry" | /telemetry |
| "learn", "extract patterns", "learn from that", "save what worked", "patterns from campaign" | /learn |
| "schedule", "recurring", "every N minutes", "cron", "set a reminder", "run periodically" | /schedule |
| "merge review", "check merges", "any conflicts", "fleet conflicts", "pending branches", "safe to merge" | /merge-review |
| "ascii diagram", "ascii art", "box diagram", "architecture diagram", "flow diagram", "draw a diagram", "text diagram", "sequence diagram" | /ascii-diagram |
| "improve", "improvement loop", "quality loop", "rubric", "score against", "run improvement", "improve citadel" | /improve |
| "evolve", "sustained improve", "improvement director", "research-driven improve", "multi-cycle improve", "run until done", "improve until ceiling", "keep improving", "hypothesis", "belief model", "scout agents" | /evolve |
| "organize", "directory structure", "folder structure", "project structure", "file organization", "where should this go", "cleanup directories" | /organize |
| "houseclean", "house clean", "disk space", "free space", "drive full", "running out of space", "clean up disk", "clean worktrees", "disk audit", "storage audit", "move to another drive", "free up space", "c drive full", "orphaned worktrees" | /houseclean |
| "daemon", "continuous", "run overnight", "keep running", "24/7", "unattended", "run autonomously", "daemon start", "daemon stop", "daemon status" | /daemon |
| "map", "index codebase", "codebase map", "structural index", "scan codebase", "map stats", "map query" | /map |
| "watch", "watch files", "watch changes", "file sentinel", "monitor files", "watch start", "watch stop", "watch scan", "marker comments", "@citadel" | /watch |
| "infra", "infrastructure", "what databases", "what systems", "docker-compose", "infra audit", "map infrastructure", "what does this connect to" | /infra-audit |
| "workspace", "multi-repo", "cross-repo", "across repos", "multiple repos", "coordinate repos", "add redis and snowflake", "split into repos" | /workspace |
If ONE skill matches with high confidence → invoke it directly. Done. High confidence = evaluator assigns ≥ 0.85 probability to exactly one skill. Below 0.85, or multiple skills above 0.70, fall through to Tier 3. If MULTIPLE skills match → carry the candidate set to Tier 3. Tier 3 disambiguates between candidates only, not from scratch. Tie-break: prefer the candidate with fewer trigger keywords.
When Tiers 0-2 don't resolve, classify across 6 dimensions:
SCOPE: single-file | single-domain | cross-domain | platform-wide
COMPLEXITY: 1 (trivial) | 2 (simple) | 3 (moderate) | 4 (complex) | 5 (campaign)
INTENT: fix | build | create | add | audit | redesign | research | improve | wire | prune
REQUIRES_PERSISTENCE: true | false (multi-session?)
REQUIRES_PARALLEL: true | false (independent sub-tasks?)
REQUIRES_TASTE: true | false (quality judgment beyond tests?)
Routing rules (first match wins):
| Condition | Route |
|---|---|
| INTENT is "create", Complexity >= 3 | /create-app |
| INTENT is "create", Complexity <= 2 | /scaffold |
| INTENT is "add", existing source files present | /create-app (Tier 5 — feature mode) |
| INTENT is "add", no existing source files | /scaffold |
| Complexity 1, single skill match | Skill directly |
| Complexity 1, no skill match | Do it yourself (direct edit) |
| Complexity 2, single domain | /marshal |
| Complexity 2-3, known skill domain | Skill, with marshal fallback |
| Complexity 3, cross-domain | /marshal |
| Complexity 3-4, requires persistence | /archon |
| Complexity 4, requires taste/judgment | /archon |
| Complexity 4-5, requires parallel | /fleet |
| Complexity 5, platform-wide | /fleet |
| Confidence < 0.7 | /marshal (safe default) |
Important: A repeated pattern complaint ("I keep doing X manually", "the agent
always makes this mistake") should route to /create-skill. A repeated pattern
is a skill waiting to be extracted.
After classification and before execution, verify the response is proportional to the input:
Downgrade triggers (apply in order):
| Condition | Action |
|---|---|
| Input < 20 words AND routed to Archon or Fleet | Downgrade to Marshal. Log: "Input too brief for campaign-level orchestration." |
| Input mentions a single file AND routed to Fleet | Downgrade to Marshal or skill. Log: "Single-file scope doesn't warrant parallel agents." |
| Estimated sessions > 5 AND user is Novice trust level | Cap at 3 sessions. Log: "Capping sessions for novice user. Run more to unlock higher budgets." |
| Routed to Daemon AND user is Novice trust level | Block. Output: "Daemon mode requires familiarity with the harness. Complete a few sessions first." |
| Estimated cost > $50 AND no explicit budget flag | Confirm with user regardless of trust level. |
Upgrade triggers:
| Condition | Action |
|---|---|
| Input complexity >= 4 AND routed to a bare skill | Suggest Marshal. "This looks complex enough for orchestration. Route to /marshal instead?" |
| Input mentions "overnight" or "continuous" AND routed to Archon | Suggest daemon. "This sounds like continuous work. Want to run it as a daemon?" (skip if Novice) |
| Input contains 2+ clearly independent tasks AND complexity >= 3 | Run Fleet auto-decomposition (see below). |
Fleet auto-decomposition — 1/2/3 confirmation prompt:
When 2+ independent tasks detected (non-overlapping scopes, complexity >= 3, not already routed to full Fleet), read consent.fleetSpawn from harness.json:
auto-allow → route directly to /fleet --quickalways-ask or null → show prompt: "These look independent — run in parallel? [1=yes 2=always 3=no]"
--quick, preference unchanged--quick, write writeConsent('fleetSpawn', 'auto-allow')always-askreadConsent/writeConsent are in hooks_src/harness-health-util.js.
Trust level: Read from harness.json trust object. Levels: novice (0-4 sessions), familiar (5-19), trusted (20+ with 2+ campaigns). trust.override takes precedence.
Log routing decision (fire-and-forget):
node .citadel/scripts/telemetry-log.cjs --event agent-complete --agent do-router --session routing --status success --meta '{"tier":N,"target":"[skill]","input_chars":M}'
Use .citadel/scripts/telemetry-log.cjs (the project-local copy). If it doesn't exist, skip logging silently — never block routing on telemetry failure.
Announce the routing decision: "Routing to [target] because [one-sentence reason]"
Invoke the target skill or orchestrator
If the target fails or the user says "wrong tool", try the next tier up. If the target is already Tier 3 (marshal fails or user explicitly escalates from a failed marshal attempt): re-route to /archon with the original input as context.
Output a grouped skill list drawn from the system reminder's available skills. Group by category (Orchestration, App Creation, Code Quality, Research & Debugging, GitHub & CI, Infrastructure, Monitoring, Utilities, Observability). For each skill, show /name — one-line description. Include a footer: "Direct invocation (/skill-name) always bypasses the router."
.planning/ does not exist: The router works without .planning/. Tiers 0, 2, and 3 are fully independent of it. Tier 1 (active-state short-circuit) reads .planning/campaigns/ and .planning/fleet/ — if those directories are absent, skip Tier 1 gracefully and fall through to Tier 2. Never crash on a missing .planning/ directory.harness.json missing: Skip the Skill Registry Check and proceed directly to Tier 0. Announce discovered skills from the filesystem if counts can be read, otherwise route from built-in keywords.--list output and a prompt to provide a direction.Disclosure: "Routing to [skill]. See that skill's contextual gates for reversibility." Reversibility: depends on routed skill — check the routed skill's reversibility Trust gates:
After routing and execution complete: