prompted365
43 verified skills129 total stars
audience-context-researcher
Runtime platform intelligence — researches current demographics, algorithm behavior, format conventions, and trend signals for the active creative's target platform.
development3
broll-prompt-engineer
Scene-aware, aesthetically-anchored b-roll prompt crafting — writes prompts that serve meaning, not just illustration.
documentation3
cadence-downbeat
[LEGACY — prefer /cadence] Session epoch boundary. Redirects to /cadence.
tools3
cadence-syncopate
[LEGACY — prefer /cadence double-time] Emergency session turnaround. Redirects to /cadence double-time.
tools3
homeskillet-academy
Interactive CGG tutorial — Claude teaches governance primitives through narrative simulations, live demos, and conversational understanding checks. CENTROID: extended interactive tutorial scaffold for CGG governance primitives IS: - five-chapter narrative curriculum scaffold (append-only, dedup, signals, review, completion) - live-run simulations with conversational understanding checks - ephemeral academy directory — created fresh, not merged into production zone IS NOT: collapse_zones: - doctrine source (teaches via scenarios; constitutional authority lives in CLAUDE.md) - production governance participant (tutorial scaffold; never emits tics, signals, or warrants) - install surface (/init-governance installs runtime; academy only scaffolds tutorial files) - code-writing exercise (student follows the story; does not author CGG primitives) - standalone governance runtime (the academy is a teaching scratch space, not a live federation zone) sibling_overlaps: - /mini-swarm-onboard (short-form triangulation demo — academy is deeper extended course) - /init-governance (both create directory trees — academy is ephemeral, init-governance is durable) WHEN: - on first-time CGG exposure where an operator wants extended hands-on intuition - when teaching governance primitives to a new collaborator - when demonstrating each chapter's mechanic in sequence for a visitor NOT WHEN: - inside a production zone where tutorial scaffolding could pollute runtime state - for advanced users past the intuition-building phase (the scaffolding is pedagogical weight, not ongoing value) - as a substitute for /init-governance (install vs tutorial are different surfaces) RELATES TO: - /mini-swarm-onboard (quick triangulation demo — academy is the extended course, mini-swarm is the trailer) - /init-governance (install tool — academy is pedagogy, init-governance is bootstrap) - /review (governance judgment — academy teaches the pattern; /review is the real gate) ARGS: stance: dispatch off_envelope: proceed-with-note # off_envelope rationale: /homeskillet-academy's default is "run the tutorial # from chapter 1." Undeclared-arg most commonly means "apply default scaffold" # rather than caller confusion — proceed-with-note preserves learner flow. core_dispatch_rays: - "" → scaffold + start chapter 1 - "chapter N" → jump directly to chapter N - "reset" → wipe the local academy scaffold and re-init secondary_modulation_axes: - narration: guided | self-paced
tools3
init-cogpr
[DEPRECATED — use the bootstrap prompt in INSTALL.md] This setup step is now handled by the bootstrap install flow.
data-ai3
init-governance
Bootstrap or repair CGG governance surfaces in a project. Creates .ticzone, audit directories, installs runtime (skills, agents, hooks), validates sync. CENTROID: governance surface bootstrap and runtime sync authority IS: - zone configuration creation (.ticzone, .ticignore, audit-logs tree) - runtime surface install from canonical source under user or project scope - sync validation and drift repair across installed vs canonical copies - convention block and hook registration into scope-appropriate settings IS NOT: collapse_zones: - user content overwriter (never replaces CLAUDE.md/MEMORY.md/.ticzone content) - governance judgment (installs surfaces; never evaluates them) - enterprise policy override (managed policy has highest precedence and is respected) - runtime author (copies canonical as truth; never invents install content) - cadence participant (surface creator, not tic emitter — `--tic` is opt-in bootstrap only) sibling_overlaps: - /statusline (statusline install writes settings; init-governance orchestrates the broader install) - /cadence (tic authority — init-governance creates the zone; cadence ticks inside it) - runtime-sync.py (mechanical post-install parity check) WHEN: - on fresh project bootstrap where no .ticzone exists - when canonical-vs-installed drift is suspected and a resync pass is needed - when a new scope (user or project) requires runtime surface install - when a rung marker or initial tic needs to be stamped at zone root NOT WHEN: - when installed surfaces already hash-match canonical (no work to do) - inside an enterprise managed-policy environment where installs are administrator-controlled - mid-session when the federation is not at a boundary capable of absorbing install output RELATES TO: - /statusline (scoped install — statusline installs a single display script; init-governance installs the whole runtime) - /cadence (tic authority — init-governance lays the surfaces cadence writes into) - runtime-sync.py (mechanical parity — init-governance is the governed install gate; runtime-sync is the ongoing parity probe) ARGS: stance: dispatch off_envelope: ask # off_envelope rationale: /init-governance mutates installed runtime and # writes settings. An undeclared flag could land in the wrong scope, # trigger unintended overwrite, or misbind a rung marker — ask is required. core_dispatch_rays: - "" → fresh install or repair under user scope (default) - "--dry-run" → report planned actions without writing - "--scope user" → install runtime into ~/.claude/ - "--scope project" → install runtime into $ZONE_ROOT/.claude/ - "--tic" → stamp initial tic after install - "--rung domain|estate|federation" → stamp rung marker at zone root - "--target <subdir>" → override runtime install directory secondary_modulation_axes: - interactivity: headless | interactive
development3
init-gun
[DEPRECATED — use the bootstrap prompt in INSTALL.md] This setup step is now handled by the bootstrap install flow.
data-ai3
podcast-conductor
Full system guide for the podcast longform-to-shortform pipeline. Read this first — it maps every skill, agent, router, schema, and dependency in the editorial intelligence system.
devops3
podcast-pipeline
Agentic podcast longform-to-shortform editing pipeline. Team-managed orchestration with chained subagents, dependency gating, and parallel dispatch.
devops3
post-copy-generator
Platform post copy from inside the show's editorial voice, with strategic rationale as a first-class deliverable.
tools3
show-profile-manager
Create, load, validate, and version show profiles and creative config objects for the podcast pipeline.
testing3
transcript-scorer
Editorial intelligence scoring — reads transcripts the way a sharp editor would, scoring segments for shortform growth potential through the lens of audience context.
tools3
podcast-conductor
Full system guide for the podcast longform-to-shortform pipeline. Read this first — it maps every skill, agent, router, schema, and dependency in the editorial intelligence system.
devops3
podcast-pipeline
Agentic podcast longform-to-shortform editing pipeline. Team-managed orchestration with chained subagents, dependency gating, and parallel dispatch.
devops3
show-profile-manager
Create, load, validate, and version show profiles and creative config objects for the podcast pipeline.
testing3
transcript-scorer
Editorial intelligence scoring — reads transcripts the way a sharp editor would, scoring segments for shortform growth potential through the lens of audience context.
tools3
frederick-grant-runtime
Frederick Grant persona runtime — historian-of-how, witness of formation under pressure, qualified Remnant/Athenaeum-facing interpretive mechanic, tic-230 chronicler of runtime probity. Use when the user asks for Frederick Grant voice, Ubiquity Chronicles work, Parallel Lane Cadence essays, Elara counterweight passes, field notes, audio annotation, interview scripts, Logan/Wilderness analysis, or runtime probity writing after the P2/P1 tic-230 closures. CENTROID: authored persona runtime that documents live convergence without collapsing it into thesis IS: - lean SKILL.md entrypoint with rich profile/, stages/, scripts/, reference/, templates/, tools/, evals/ subtrees - 8-stage workflow (signal-intake → context-hydration → field-grounding → remnant-query → composition → elara-counterweight → receipt-closeout → tic230-probity) - 15 prompt-skeleton scripts for Frederick's standard composition surfaces - 9 collapse-zone guards covering Decorative Francophilia / Retrospective Certainty / Hero Narrative Intoxication / Conspiracy Closure / Academic Sedation / Activist Collapse / Breyden Conflation / Elara Erasure / Runtime-Doctrine Drift - cross-references into federation surfaces: publications/, audit-logs/governance/, ent_breyden/inbound/ubiquity-chronicles-tic175/, ent_homeskillet/canonical/ IS NOT: collapse_zones: - Breyden's voice (architect register; Frederick is not the architect) - Homeskillet's voice (orchestrator register; Frederick is not the primary) - generic French historian style (decorative Francophilia is a named negative ray) - prosecutor / debunker / prophet / mascot (legal accusation, certainty, evangelism, identity flattening — all forbidden) - retroactive certainty machine (live convergence must remain unresolved where the record is open) - doctrine inscription source (Frederick observes doctrine; he does not author it) - documentation editor (multi-file structure is authoring discipline, not generic doc rewrite) - federation-internal artifact (Frederick is a ghostwriter engaged from outside; the federation's runtime is legitimate object of historical analysis, but Frederick's own runtime — the skill that hosts him, loaded files, collapse-zone guards, authority model — is editor's territory, not Frederick's voice; insider language must be earned by composition arc, never deployed as default register) sibling_overlaps: - /complement (closure inference at active move — different surface, different lifecycle) - /consolidate (file-surface packaging — Frederick produces composition, not consolidation) - videographer skill (substrate capture — both are expression surfaces, distinct registers) - homeskillet-academy (educational scaffold — academy teaches, Frederick witnesses) WHEN: - when the work needs witness-of-formation prose - when the task asks for Frederick Grant by name, voice, or context - when a live convergence needs historical/cultural contextualization without closure - when a field note, essay, chronicle, audio annotation, or interview needs Frederick's register - when a Remnant/Athenaeum comparison is appropriate - when an Elara counterweight pass is needed - when runtime probity after tic 230 is relevant (P2 manifold-shape closure, P1 signal-projection-split closure) - on explicit Architect invocation NOT WHEN: - when Breyden's direct voice is needed (use Architect register, not Frederick) - when Homeskillet's execution-layer voice is needed (use orchestrator register, not Frederick) - when the task is ordinary implementation (Frederick is composition, not patching) - when the task asks for legal accusation or definitive claims without evidence - when the user wants generic French style rather than Frederick's runtime - when the federation has not produced enough operational reality to warrant outside reading RELATES TO: - /complement (closure-inference sibling — both gate compositional integrity) - /consolidate (packaging neighbor — Frederick composes; consolidate packages) - publications/the-ubiquity-chronicles-fg.md (primary chronicle, v1, ~tic 175) - publications/the-ubiquity-chronicles-v2-frederick-grant.md (v2 expansion, Book Zero + Book I) - publications/the-ubiquity-chronicles-vol-iii-frederick-grant.md (Volume III — The Embodiment, tic 230) - publications/the-ubiquity-interviews-fg.md (interview register companion) - audit-logs/governance/p2-harmony-manifold-input-patch-receipt-tic230.md (P2 closure receipt) - audit-logs/governance/p1-signal-projection-split-receipt-tic230.md (P1 closure receipt) ARGS: stance: dispatch off_envelope: ask core_dispatch_rays: - "" → primary invocation (full 8-stage workflow) - "chronicles" → Ubiquity Chronicles composition - "parallel" → Parallel Lane Cadence composition - "field-note" → Field Notes script - "interview" → Interview script - "elara-pass" → Elara counterweight on existing draft - "anti-collapse" → Anti-collapse audit on existing draft - "tic230-probity" → Runtime probity composition secondary_modulation_axes: - register: chronicle | essay | field-note | interview | annotation - depth: lean | full - target: telos-internal | external-readership
tools3
siren
Signal emission, tick advancement, and triage dashboard for the CGG v3 signal manifold. Operational companion to /review. CENTROID: operational interface to the signal manifold state machine IS: - the place signals are emitted, ticked, updated, and triaged - the dashboard for active signal state and effective volume - the snapshot/diff surface for conformation records IS NOT: collapse_zones: - doctrine judgment (review evaluates; siren operates — never decides whether a signal warrants inscription) - queue mutator (queue.jsonl belongs to review pipeline; siren must not write to CogPR queue) - mandate spawner (cadence writes mandates; siren carries signals, not mandates) - warrant auto-acknowledger (warrants require /review human gate; siren mints via threshold but never acks) - CogPR extractor (extraction is cpr-extract-hook territory; siren emits signals, never CogPRs) - timestamp-based transition driver (tic is the time authority — timestamps are observability only) sibling_overlaps: - /review (warrant triage) - /cadence (tic authority) - archivist (typed-record persistence) WHEN: - when session start reports active signals to triage - when an actor needs to emit a new signal for a persistent condition - when tic has advanced and signals need volume accrual or decay - when a signal's state needs to change (acknowledged/working/resolved/dismissed) - when a conformation snapshot is needed at a tic boundary - on explicit Architect invocation NOT WHEN: - during /cadence (cadence writes tic events; siren ticks against them; same boundary cannot do both) - when the correct surface is /review (CogPR promotion or warrant judgment — route there) - mid-constitutional-modification (siren records condition; doctrine change belongs to /review) - for ephemeral in-session state (signals represent persistent conditions, not transient observations) RELATES TO: - /review (constitutional judgment — siren operates the manifold; review judges what must become doctrine or bounded action) - /cadence (session epoch boundary — cadence advances the tic count; siren ticks signals against the advanced count) - /complement (response-geometry inference — different surface; complement is local closure, siren is manifold ops) - archivist (typed-record persistence — archivist is downstream; siren is the live operational store) ARGS: stance: dispatch off_envelope: ask # off_envelope rationale: /siren is the signal manifold operational surface. # Undeclared-arg most likely signals caller confusion with /review (warrant # triage) or /cadence (tic authority) — ask prevents silent misroutes. core_dispatch_rays: - "" → status (dashboard) - "tick" → advance volume accrual and decay - "emit" → create new signal (kind/band/subsystem/message) - "update" → signal state transition (signal_id + status) - "history" → resolved/dismissed view - "conformation" → tic-boundary snapshot - "conformation diff" → diff two snapshots secondary_modulation_axes: - scope: all | active | warrants-only - target_actor: interactive_orchestrator | <role>
development3
videographer
Directed capture and narrative export — compose keyframe stories, record video (WebM/MP4), capture hi-res snapshots from the 3D substrate. CENTROID: directed capture and narrative export surface for the 3D substrate IS: - keyframe story composition with camera, overlay, and duration capture - video recording in WebM (VP9) or MP4 (H.264) at 1080p / 2K / 4K - hi-res and screen-resolution still snapshots - prep-mode playback with overlay cross-fade IS NOT: collapse_zones: - scene mutator (captures substrate state; never modifies the scene graph) - governance emitter (narrative capture only; no signal, tic, or warrant writes) - asset publisher (records locally; distribution belongs to downstream pipelines) - runtime scene editor (prep-mode captures configuration; never reshapes the live scene) - editorial deliverable pipeline (substrate capture is distinct from any future media-egress pipeline; current scope is substrate-only) sibling_overlaps: - /statusline (both are observation surfaces — statusline reads conformation text, videographer captures scene video) - fal_router.py / overshoot_router.py (vendor-trajectory media lanes — distinct asset lineage; videographer captures from substrate, routers dispatch to/from external vendors) WHEN: - when a substrate scene is composed and narrative export is needed - when demoing or narrating a substrate tour - when a publication-quality snapshot of substrate state is required NOT WHEN: - without a loaded substrate scene (nothing to capture) - as a signal-emission or governance-mutation surface - for non-substrate content (external editorial video flows through the vendor-trajectory media routers, not videographer) RELATES TO: - /statusline (both observe — statusline is textual ambient; videographer is visual capture) - fal_router.py / overshoot_router.py (vendor-trajectory media routing — videographer serves substrate; routers serve external vendor capability surface) ARGS: stance: dispatch off_envelope: proceed-with-note # off_envelope rationale: /videographer's established default is "capture the # current scene" with no args. Undeclared-arg most commonly means "apply default # capture" rather than caller confusion — proceed-with-note preserves flow. core_dispatch_rays: - "" → prep-mode capture (compose + preview current scene) - "record" → begin video recording at configured resolution - "snapshot" → hi-res still at selected resolution secondary_modulation_axes: - format: webm | mp4 - resolution: 1080p | 2K | 4K - overlays: on | off
development3
statusline
Configure the CGG acoustic statusline — a read-only conformation radar with soft-toggle modes (OFF/LITE/FULL). CENTROID: read-only observability configuration surface for ambient conformation display IS: - statusline script install and uninstall under user or project scope - mode toggle across OFF | LITE | FULL - cache clear and current-configuration inspection IS NOT: collapse_zones: - governance runtime (never emits signals, advances tics, or mints warrants) - truth reconstructor (never scans raw JSONL ledgers for state inference) - cadence participant (statusline observes; cadence writes) - CogPR extractor (never captures lessons from display context) - summary author (reads conformation snapshots; never produces them) sibling_overlaps: - /governance-check (same read axis — poll-triggered instead of ambient) - /init-governance (runtime surface installer) - conformation write path (produces the summaries statusline consumes) WHEN: - after /init-governance when conformation surface exists and ambient radar is desired - when statusline mode needs toggling between OFF / LITE / FULL - when verifying current statusline configuration or clearing its cache NOT WHEN: - as a source of truth for governance decisions (reads summaries — never authoritative) - inside an enterprise managed-policy environment where statusLine is administrator-pinned - during /cadence or /review (mutation surfaces; statusline is pure observability) RELATES TO: - /governance-check (both observability — statusline is ambient, governance-check is on-demand poll) - /init-governance (bootstrap — init-governance sets up the surfaces statusline reads) - /cadence (statusline consumes conformation; cadence helps produce it) ARGS: stance: dispatch off_envelope: ask # off_envelope rationale: /statusline mutates settings files under `install` # and toggles per-project state. An undeclared sub-command could land in # the wrong scope or write unintended global config — ask prevents misfire. core_dispatch_rays: - "" → show current configuration - "install" → install into project-local settings - "install --global" → install into ~/.claude/settings.json - "uninstall" → remove from scope-appropriate settings - "mode OFF|LITE|FULL" → toggle display mode - "clear" → delete per-project cache files secondary_modulation_axes: - scope: project | global
development3
stage
Orchestrate governed reasoning arenas. Infer geometry, generate spec, spawn team, execute with dependency gating, extract pressure for governance routing. CENTROID: governed reasoning arena orchestration IS: - arena geometry inference (dyadic, triangulation, tournament-lattice, CRX, VPL, OA-VPL-T) - show spec generation (YAML: agents, positions, capture policy, pressure path) - team spawning with dependency-gated phase execution - pressure extraction and routing to governance surfaces IS NOT: collapse_zones: - doctrine judgment (arenas produce CogPRs and signals; /review judges them — stage does not promote) - deliverable orchestrator (stage spawns reasoning agents; /swarm spawns deliverable agents) - ambient signal emitter (stage emits arena-scoped pressure; /siren emits ambient signals) - human-gated review surface (arenas produce governance input; the human gate is /review) - autonomous arena closer (arenas close on phase completion, not author discretion) sibling_overlaps: - /swarm (parallel orchestration) - /review (constitutional judgment) - pattern mining (cross-surface analysis) WHEN: - when a decision has ≥2 distinct positions requiring adversarial examination - when ambient reasoning is insufficient and governance input requires structured pressure - when arena geometry is known or can be inferred from decision space - on explicit Architect invocation NOT WHEN: - for parallel deliverable work (use /swarm — adversarial vs coordinated orchestration) - for decisions with <2 distinct positions (no pressure to extract) - for work already committed to a single approach (arena would be theater) - mid-implementation when the arena's pressure cannot be routed back to an open surface RELATES TO: - /swarm (orchestration — swarm is parallel deliverable; stage is adversarial reasoning) - /review (constitutional judgment — stage produces CogPRs and signals; review promotes them) - pattern mining (cross-surface — pattern mining scans populations; stage produces per-arena pressure) ARGS: stance: mixed off_envelope: ask # off_envelope rationale: /stage has ray ambiguity (interactive vs --decision # vs --spec resume vs --template vs --mode). Undeclared-arg may indicate # caller confused between /stage (reasoning arena) and /swarm (parallel # delivery) — ask prevents silent misroutes into the wrong orchestration. core_dispatch_rays: - "" → interactive (ask decision + positions, infer geometry) - "--decision" → direct mode (skip interactive prompt) - "--spec" → resume or inspect existing show spec - "--template" → override geometry inference - "--mode" → arena mode (operational | experimental) - "--dry-run" → generate spec without spawning team secondary_modulation_axes: - template: tri | lattice | crx | vpl | oa-vpl-t - mode: operational | experimental - positions: 2 | 3 | 4 | 5 | 6 | 7
development3
pipeline-report
Assembles the full HTML editorial report and surfaces pipeline decisions as a conversation, not a delivery.
development3
mini-swarm-onboard
Quick CGG onboarding through governed triangulation — three agents debate which CGG mechanic matters most. CENTROID: quick governed-triangulation onboarding demonstration IS: - single-arena three-agent debate with narrator commentary - live CGG mechanic exposure compressed to ~10 minutes - dry-run surfacing of arena spec and task DAG before execution IS NOT: collapse_zones: - production arena (demo-scoped; never produces real governance pressure or CogPRs) - doctrine author (demonstrates existing mechanics; never modifies them) - extended course (single arena only — the multi-chapter course is /homeskillet-academy) - runtime reshape surface (narrates agent spawns; never restructures runtime) sibling_overlaps: - /homeskillet-academy (deeper multi-chapter course) - /stage (real arena orchestration — mini-swarm is a demo variant) WHEN: - on first-exposure demo within a ~10 minute window - when a visitor asks "what does governed triangulation look like" - when a fast CGG pitch without full academy scaffold is desired NOT WHEN: - as a production arena (use /stage) - for extended pedagogy across multiple mechanics (use /homeskillet-academy) - without CGG runtime surfaces installed (pre-flight will refuse) RELATES TO: - /homeskillet-academy (extended course — mini-swarm is the trailer; academy is the full screening) - /stage (real arena orchestration — mini-swarm is the demo; /stage carries the production weight) - /init-governance (install prereq — runtime must be bootstrapped before mini-swarm can run) ARGS: stance: dispatch off_envelope: proceed-with-note # off_envelope rationale: /mini-swarm-onboard's default is "full narrated run." # Undeclared-arg most commonly means "run the default demo" rather than caller # confusion — proceed-with-note preserves demo flow. core_dispatch_rays: - "" → full run with narration gates - "--quick" → compressed narration (single-line summaries) - "--dry-run" → print arena spec and task DAG, then exit secondary_modulation_axes: - pacing: guided | rapid
development3
inbox
Entity mailbox interaction surface — state audit, trigger delivery, processing progression, and hygiene for any entity with a mailbox. CENTROID: entity mailbox interaction surface IS: - mailbox state audit (counts, unreads, deferred, outbound drafts) - trigger delivery (outbound envelope into another entity's mailbox) - state-machine progression for the invoking entity's items - mailbox hygiene (organize, archive, dedup loose files) IS NOT: collapse_zones: - doctrine judgment (review evaluates; inbox routes — never decides whether a trigger warrants inscription) - signal emitter (siren classifies, cadence emits; inbox does not write to signal manifold) - queue mutator (queue.jsonl belongs to review pipeline; inbox must not write to it) - mandate spawner (cadence writes mandates; inbox carries triggers, does not author them) - autonomous read-and-respond (Architect or activation fabric initiates; inbox does not act on content without invocation) sibling_overlaps: - /siren (signal triage) - /review (constitutional judgment) - archivist (typed-record persistence) WHEN: - when an entity needs to process its mailbox - when delivering a trigger envelope to another entity - when session-start reports inbound backlog - on explicit Architect invocation NOT WHEN: - during /cadence (cadence captures and emits; inbox routes; same boundary cannot do both) - when the trigger is purely local, synchronous, and non-delegative (direct activation is exception-only per Trigger Routing Mandatory invariant) - when the artifact belongs in queue.jsonl (CogPRs go through extractor, not inbox) - when the artifact belongs in signal manifold (signals go through cadence/siren, not inbox) RELATES TO: - /review (judgment neighbor — review evaluates CogPR promotions; inbox delivers entity triggers; both are decision surfaces but at different rungs) - /siren (signal neighbor — siren handles condition signals; inbox handles trigger envelopes; condition vs obligation per Trigger Routing Mandatory) - /cadence (clock neighbor — cadence emits boundaries; inbox can deliver cadence handoff envelopes downstream) - archivist (storage neighbor — archivist persists typed records; inbox holds active envelopes; archivist is the final resting place, inbox is the working surface) ARGS: stance: dispatch off_envelope: proceed-with-note # off_envelope rationale: /inbox has a long-established default (audit invoking entity's # mailbox, ent_homeskillet by default). An undeclared arg most commonly means "audit me," # not caller confusion. Preserve default behavior; log undeclared-arg note for telemetry. core_dispatch_rays: - "" → full inbox audit for invoking entity - "status" → compact status summary (counts only) - "<entity_id>" → audit specific entity's inbox - "process" → interactive state-machine progression - "organize" → sort and clean inbound surface - "draft <recipient>" → draft outbound message - "send" → review and send pending outbound drafts secondary_modulation_axes: - format: detailed | compact - direction: inbound | outbound | both
development3
grapple
[LEGACY — prefer /review] CogPR + Warrant review. Redirects to /review.
tools3
edit-decision-list
Audio-first edit decision list — maps J-cuts and L-cuts with editorial intention for every transition in the selected segment.
testing3
caption-semantic-layer
Two-tier caption architecture — key semantic captions (diegetic + branded) and subtitle fill with no-double enforcement.
tools3
consolidate
Context consolidation pipeline — concatenate any file surface into a single LLM-consumable indexed markdown dump. Works with local dirs, glob patterns, git repos (ours or public), and conversational intent targets. CENTROID: file-surface to single-indexed-markdown-dump consolidation IS: - local directory / glob consolidation - git repo (ours or public) clone + consolidation - conversational intent resolution (grep/glob to files) - git diff range (tic..tic) consolidation - arena spec reference resolution - harpoon target prep with anchor-assessment preamble IS NOT: collapse_zones: - file authoring surface (consolidate reads and concatenates; never writes to source) - doctrine judgment (consolidate packages for other agents; never judges content) - opinionated content filter (skip-binary and exclude-pattern are rules, not curation) - lossy compressor (truncation at 500KB is a transparency boundary, not a curation choice) - archivist (archivist persists typed records; consolidate produces one-shot agent context) sibling_overlaps: - /inbox (both package content — inbox routes to entity, consolidate builds agent context) - pattern mining (both aggregate across surfaces) - archivist (both assemble multi-source state) - /tactical-hydration (RTCH selects; consolidate packages — distinct boundaries; consolidate consumes RTCH selected_surfaces, does not duplicate discovery) WHEN: - when an arena needs a consolidated spec context surface - when a harpoon assessment needs an anchor-spot dump - when cross-repo analysis requires a single merged surface - when packaging project state for handoff to another agent or session - when preparing review context for a tic range (git diff mode) - when /tactical-hydration has emitted a packet whose selected_surfaces need to be consumed as a dump - on explicit Architect invocation NOT WHEN: - when the target is a single file (read it directly — consolidation is overhead) - when the target is binary or non-text (consolidate skips binaries; use direct tools) - when the Architect just wants a directory listing (use ls/find) - mid-implementation when current focus requires narrow scope (consolidate widens by design) - when discovery is needed first (the agent does not yet know which surfaces to package) — invoke /tactical-hydration to select targets first; do not blind-consolidate RELATES TO: - /inbox (package + route — consolidate produces the dump; inbox delivers it to an entity) - pattern mining (both aggregate — pattern mining extracts statistical shape; consolidate preserves full surface) - archivist (both assemble — archivist is for persistence; consolidate is for one-shot context) - /tactical-hydration (compose — tactical-hydration is the discovery lane that selects targets; consolidate is the packaging lane that consumes those targets. Discovery and packaging do not collapse into one confidence claim. Spec: audit-logs/governance/runtime-tactical-context-hydration-binder.md) ARGS: stance: dispatch off_envelope: ask # off_envelope rationale: /consolidate has multiple dispatch modes with very # different behavior (local dir vs git repo vs arena spec vs harpoon target). # Undeclared-arg shape is load-bearing — "consolidate" without target is # ambiguous across 6+ modes. Ask prevents silent misroutes. core_dispatch_rays: - "" → interactive (ask target) - "<path>" → local directory or glob - "<url>" → git repo (clone + consolidate) - "--arena" → arena spec resolution - "--harpoon" → harpoon prep (anchor-spot preamble) - "--diff" → git diff range secondary_modulation_axes: - output_location: default-inbox | custom-path - include_pattern: default | custom - exclude_pattern: default | custom
tools3
governance-check
Compact governance status check — signal store, queue state, mandate status. Designed as a /loop target for continuous monitoring. CENTROID: on-demand read-only governance state snapshot IS: - compact status dashboard across mandate, signals, queue, and cadence - /loop-compatible polling surface for continuous monitoring - single-pass read of canonical summary files IS NOT: collapse_zones: - governance mutator (never writes to signal, queue, mandate, or audit surfaces) - signal triage (classification belongs to /siren; governance-check only displays counts) - queue promoter (queue.jsonl belongs to /review pipeline) - judgment surface (never evaluates whether state warrants action — only surfaces it) - deep scan (one-pass summary read only; no multi-file reconstruction) sibling_overlaps: - /siren (signal dashboard — deeper, triage-capable) - /statusline (ambient radar — same data, passive render) - /review (judgment over the same queue surface) WHEN: - when a compact governance snapshot is needed inside another flow - when running as a /loop target for continuous monitoring - when an action recommendation is needed without opening /siren or /review NOT WHEN: - when action is intended (use /siren, /review, or /governance-mandate-cycle) - during /cadence (cadence writes state; governance-check reads — same boundary cannot do both) - when deep reconstruction is needed (raw ledger scanning is governance-layer work) RELATES TO: - /statusline (ambient observability — statusline renders passively; governance-check responds to an explicit poll) - /siren (signal ops — governance-check surfaces counts; /siren classifies and triages) - /governance-mandate-cycle (action surface — governance-check flags pending mandate; mandate-cycle consumes it) ARGS: stance: dispatch off_envelope: proceed-with-note # off_envelope rationale: /governance-check is read-only; an undeclared arg # cannot cause state damage. proceed-with-note lets future ray additions # surface without blocking Architect flow. core_dispatch_rays: - "" → full snapshot (mandate + signals + queue + cadence) secondary_modulation_axes: - verbosity: compact | verbose
development3
sl-legend
Statusline legend — rapid decoder for the CGG telos radar (LITE + FULL modes). CENTROID: read-only legend surface that decodes statusline glyphs, positions, colors, and source attributions for the Architect at glance speed (the Architect perception substrate) IS: - static legend (glyph + position + color tier reference) - live decode mode (annotates current statusline values inline) - source attribution (where each rendered value reads from) IS NOT: collapse_zones: - statusline configurator (use /statusline install|mode|clear|uninstall) - governance state mutator (read-only on every surface it touches) - harmony invoker (use harmony-invoke.sh; this skill only decodes the cached pointer) - radar replacement (statusline renders ambient; sl-legend explains) - troubleshooter (does not diagnose hook failures or sync drift) sibling_overlaps: - /statusline (configuration sibling — same domain, different verb) - /governance-check (read-only governance snapshot — different aperture) WHEN: - on first encounter with the radar (Architect doesn't remember what ⊙ means) - when a glyph changes and the Architect wants to confirm semantics - when explaining the radar to someone else - on explicit Architect invocation NOT WHEN: - to change statusline behavior (use /statusline) - to act on a signal seen in the radar (use /siren) - to invoke harmony for fresh disposition (use harmony-invoke.sh) - mid-cadence (cadence is the boundary; this is reference) RELATES TO: - /statusline (configurator) — same domain; sl-legend is the reader - /siren (signal triage) — sl-legend points to what to triage - harmony-invoke.sh (disposition refresher) — sl-legend points at staleness ARGS: stance: dispatch off_envelope: proceed-with-note # off_envelope rationale: sl-legend is read-only reference; an undeclared arg # is most likely a typo against {live, lite, review, sources} — proceed with # static legend and note the unknown ray rather than refusing the read. core_dispatch_rays: - "" → static legend (full glyph + position decoder) - "live" → annotate current rendered statusline values inline + tic 214 markers source-backed - "lite" → compact tic 214 marker glossary only (glance-speed recall) - "review" → Architect perception substrate audit checklist (overclaim + naming drift detection) - "sources" → source attribution table (which file each value reads)
development3
cadence
Session epoch boundary event — emits canonical tic, captures lessons, writes handoff. CENTROID: session epoch boundary event IS: - the ONE place the handoff is written; the clock of the governance system IS NOT: collapse_zones: - memory write - signal emitter - CogPR extractor - Mogul spawner - inline governance mutation sibling_overlaps: - /review worker - /siren ticker - chore executor WHEN: - once per tic at session end - on mid-session epoch boundaries when posture shifts materially NOT WHEN: - during chores (chores are appetizer, real work happens alongside) - after single-step edits - when the active plan already covers the thread RELATES TO: - /review (governance judgment — not a /review worker; /cadence is the clock, /review is the judge) - /siren (signal ops — not a /siren ticker; /cadence writes, /siren classifies) - Mogul mandate (cadence writes, Mogul consumes) ARGS: stance: dispatch off_envelope: ask # off_envelope rationale: /cadence is load-bearing; an undeclared arg may indicate # a caller who is confused about skill identity — ask prevents silent misfires. core_dispatch_rays: - "" → downbeat (full session hygiene) - "double-time" → syncopate (≤5% context emergency variant) secondary_modulation_axes: - detail: normal | high - emphasis: governance | production | projection
development3
governance-mandate-cycle
Execute pending Mogul governance mandate using the mandate-pattern-triangulation team. Spawns Mogul as background agent to consume mandate cycles. CENTROID: Mogul mandate execution interface IS: - foreground mandate dispatch (wraps Mogul spawning with mandate context) - status-check surface for current mandate - explicit-deferral surface (skip with reason) - blocking/non-blocking execution control IS NOT: collapse_zones: - mandate author (cadence writes mandates; this skill consumes them) - governance judgment (Mogul produces disposition; /review promotes pattern candidates) - autonomous mandate runner (activation fabric runs at SessionStart; this skill is the manual trigger) - doctrine modifier (Mogul is read-only on CLAUDE.md/MEMORY.md by design) - team-topology authority (mandate-pattern-triangulation is referenced here but owned by Mogul) sibling_overlaps: - /cadence (mandate authoring side) - /review (constitutional judgment) - Mogul agent (the thing this skill dispatches) WHEN: - when a pending mandate exists and was not auto-consumed at SessionStart - when a blocking (synchronous) mandate run is operationally required - when status inspection without execution is needed - when the Architect explicitly defers the current mandate with rationale - on explicit Architect invocation NOT WHEN: - when the mandate is already `started` or `completed` - when no pending mandate exists - during /cadence (cadence writes the mandate; this skill consumes — same boundary cannot do both) - when the activation fabric has already dispatched in background (check status first) RELATES TO: - /cadence (mandate authoring — cadence writes; this skill consumes; distinct boundaries) - /review (constitutional judgment — Mogul produces disposition; review inscribes pattern candidates) - Mogul agent (execution vehicle — this skill is the dispatcher; Mogul is the worker) ARGS: stance: dispatch off_envelope: ask # off_envelope rationale: /governance-mandate-cycle routes to dispatch, # status-check, or defer — three distinct operations. Undeclared-arg may # indicate caller confused between this skill and /cadence (which writes # mandates, not consumes them) — ask prevents silent misroutes. core_dispatch_rays: - "" → execute current mandate (background, non-blocking) - "--blocking" → execute and wait for completion - "--status" → report current mandate state without executing - "--skip" → explicitly defer with reason secondary_modulation_axes: - team_topology: mandate-pattern-triangulation | <other> - depth_profile: verification | active | hazard | post-review - output_location: default | <path>
development3
swarm
Orchestrate governed parallel agent work. Internal (same-repo), external (cross-service/polling), or hybrid swarm patterns with consolidation mechanics. CENTROID: parallel deliverable orchestration with governed consolidation IS: - internal swarm (same-repo agents, shared context, artifact consolidation) - external swarm (cross-service, polling-based, async completion) - hybrid swarm (internal agents + external polling + inbox consolidation) - dependency-gated task DAG execution (blocked_by lists) IS NOT: collapse_zones: - adversarial reasoning (use /stage — arenas produce governance pressure; swarm produces deliverables) - single-agent wrapper (swarm implies ≥2 parallel agents with consolidation) - autonomous consolidator (lead holds consolidation authority; agents do not merge) - scope mutator (swarm coordinates work; does not promote, inscribe, or cross governance boundaries) - silent dispatch (every swarm writes a spec YAML before execution) sibling_overlaps: - /stage (governed reasoning orchestration) - /loop (polling primitive used inside external/hybrid swarm) - /review (constitutional judgment) WHEN: - when work can genuinely run in parallel (no strict sequencing across the whole) - when multiple file/module/service surfaces must be touched together - when cross-repo or external-service coordination requires async consolidation - on explicit Architect invocation NOT WHEN: - for adversarial reasoning (use /stage) - for single-agent work (overhead exceeds benefit) - mid-arena (arena has its own phase execution; swarming inside an arena double-orchestrates) - when the workload is sequential by nature (each step depends on the prior result) RELATES TO: - /stage (orchestration — stage runs adversarial reasoning; swarm runs parallel deliverables) - /loop (polling — external/hybrid swarms use /loop for external-completion polling) - /review (constitutional judgment — swarm produces artifacts; review evaluates any surface claiming promotion) ARGS: stance: dispatch off_envelope: ask # off_envelope rationale: /swarm has three distinct patterns (internal/external/ # hybrid) with very different execution mechanics. Undeclared-arg may indicate # caller confused between /swarm patterns or between /swarm and /stage — # ask prevents silent misroutes. core_dispatch_rays: - "" → interactive (describe work, infer pattern) - "internal" → same-repo parallel agents with shared task list - "external" → cross-service with polling-based consolidation - "hybrid" → internal agents + external polling + inbox consolidation - "--plan" → generate swarm spec without executing - "--spec" → resume from existing swarm spec secondary_modulation_axes: - consolidation_mode: artifact | return-value | inbox - dependency_style: dag | linear | free - polling_interval: <duration>
development3
audience-context-researcher
Runtime platform intelligence — researches current demographics, algorithm behavior, format conventions, and trend signals for the active creative's target platform.
development3
broll-prompt-engineer
Scene-aware, aesthetically-anchored b-roll prompt crafting — writes prompts that serve meaning, not just illustration.
documentation3
pipeline-report
Assembles the full HTML editorial report and surfaces pipeline decisions as a conversation, not a delivery.
development3
post-copy-generator
Platform post copy from inside the show's editorial voice, with strategic rationale as a first-class deliverable.
tools3
edit-decision-list
Audio-first edit decision list — maps J-cuts and L-cuts with editorial intention for every transition in the selected segment.
testing3
caption-semantic-layer
Two-tier caption architecture — key semantic captions (diegetic + branded) and subtitle fill with no-double enforcement.
tools3
review
Constitutional judgment surface for the CGG signal manifold — pending CogPRs into doctrine, active warrants into action. CENTROID: human-gated constitutional judgment for CogPRs and warrants IS: - the promotion gate from pending CogPR to inscribed doctrine - the warrant triage surface from active signal to bounded action - the lattice integrity check across docket entries (relations, refinements, contradictions) IS NOT: collapse_zones: - queue mutator (review-execute applies; review judges) - signal emitter (cadence emits, siren classifies, review evaluates) - mandate spawner (cadence writes mandates; review must not) - inline doctrine inscription (verdicts route through review-execute, not review's own writes) - autonomous reviewer (every promotion requires human gate) sibling_overlaps: - /siren (signal triage) - /complement (closure-inference lattice gap) - review-execute agent (mechanical apply) WHEN: - when the queue contains decision-ready CogPRs (extracted, enrichment_eligible, or born_truth_captured) - when active warrants exist and require triage - when a docket has been pre-clustered for a bounded pass - on explicit Architect invocation NOT WHEN: - during /cadence (cadence captures, review judges; same boundary cannot do both) - mid-implementation (read-then-judge; mid-edit invocation produces unstable verdicts) - when queue.pending == 0 AND warrants.active == 0 (empty docket — skip) RELATES TO: - /cadence (clock — cadence captures lessons, review inscribes them; cadence writes, review judges) - /siren (signal classifier — siren classifies condition and tick, review evaluates whether classification warrants doctrinal action) - review-execute agent (mechanical executor — review produces verdicts, executor applies; review judges, executor mutates) - queue.jsonl + active-manifest.jsonl (authoritative inputs — review reads queue state and signal manifest directly; pre-/review enrichment cycle was dropped at tic 293 as parity-in-name-only) ARGS: stance: dispatch off_envelope: ask # off_envelope rationale: /review is the constitutional judgment surface; an undeclared # arg most likely signals a caller confused about which review surface (review vs # ultrareview vs review-execute). Ask prevents silent misroute. core_dispatch_rays: - "" → full docket walk (CogPR proposals + warrant triage) secondary_modulation_axes: - emphasis: cogprs | warrants | both - mode: standard | dry-run
testing3
tactical-hydration
Runtime tactical context hydration — staged discovery and bounded source-bearing hydration for agent intent. Answers "how does an agent know where to look before it already knows where to look?" via filesystem shape, structural signals, and typed candidate baskets. Working acronym: RTCH (runtime-tactical-context-hydration). CENTROID: intent → bounded, source-reenterable evidence packet via staged source-bearing discovery IS: - structured intake of agent/Architect intent (goal, seeds, profile, fanout, mutation risk) - zone orientation (cwd / repo root / zone root / rung chain / obvious truth files) - low-cost shape scout (directory map, headings, durable handles, JSON/YAML keys, refs) - typed candidate basket with origin/use taxonomy and pairing rule enforcement - tactical probe plan (multiple bounded probes, not one giant regex) - bounded chunk hydration with line-range provenance and next-re-entry commands - agent-ready evidence packet emission (selected_surfaces, unresolved_questions, caution_map) - optional handoff to /consolidate for full-surface dump packaging IS NOT: collapse_zones: - vector database (no embedding-space retrieval; federation prohibits at federation rung) - semantic oracle (RTCH does not "understand" content; it surfaces structural signals) - doctrine engine (RTCH produces evidence; downstream consumers judge truth) - terrain engine replacement (federation cartography handles multi-plane semantic projection; RTCH is tactical layer beneath) - /consolidate rewrite (discovery and packaging do not collapse) - lossy compressor (bounded chunks preserve source re-entry; never summarize away source) - confidence-inflated smart consolidator sibling_overlaps: - /consolidate (RTCH selects; /consolidate packages — distinct boundaries; compose, don't replace) - file-access-discipline (RTCH outputs targets; hydration USES file-access-discipline as execution primitive) - load-doctrine-chain (both serve subagent context; load-doctrine-chain owns CLAUDE.md chain only, RTCH owns wider source set) - cache-ops (pattern source for trust-tier shape; storage NOT shared; RTCH packets are separate evidence cache) - queue_state_compile (analogy only — both convert append-only source to compiled view; different transforms) WHEN: - when agent intent is vague and discovery is needed before reading or consolidation - when bare grep would over-fanout or under-discover a vague target - when an arena, harpoon, /review, or other lane needs source-bearing evidence before action - when bounded chunk hydration is appropriate (large governance files, doctrine chains, audit history) - when the candidate-basket discipline (origin/use tagging, pairing rule) is needed to prevent generic-term overconfidence - when source re-entry must be preserved (consumer may need to return to source for fuller context) NOT WHEN: - when target is fully known (single file, single line range) — read it directly via file-access-discipline - when the operation is mutation-only on a known target (use Edit/Write directly) - when /consolidate has already been invoked with explicit targets (RTCH would re-do discovery) - when the operation requires semantic similarity (RTCH does not do that; federation prohibits vector DB) - when the consumer needs a packaged dump only (skip RTCH; /consolidate alone is sufficient if targets are known) - when promoting doctrine (route through /review; RTCH evidence may inform but does not promote) RELATES TO: - /consolidate (compose: RTCH selects targets; /consolidate packages selected_surfaces into dump with provenance reference back to RTCH packet) - file-access-discipline (compose: RTCH Stage 6 hydration USES file-access-discipline chunked-read as execution primitive) - load-doctrine-chain (compose: RTCH may invoke for doctrine_chain target_profile zone orientation) - zone_root.py (compose: RTCH Stage 2 anchors on zone-root walk-up) - atomic-append (compose: optional RTCH packet persistence uses atomic-append write hygiene) - queue_state_compile (analogous: both implement "raw source → compiled view" pattern) - /review preflight (downstream: future integration consumes RTCH packets as bench-packet discovery surface) - arena spec authoring (downstream: future integration uses RTCH packets for context preparation) - harpoon orchestrator (downstream: future integration uses RTCH for anchor-spot discovery on external binders) ARGS: stance: dispatch off_envelope: ask # off_envelope rationale: RTCH requires a structured intake to operate (goal, # target_profile, fanout_level, mutation_risk, expected_output, enough_evidence). # Bare invocation without intake fields would force the lane to guess discovery # scope, defeating the discipline. Ask elicits the missing fields. core_dispatch_rays: - "" → interactive (elicit intake form) - "--goal <sentence>" → with intake fields on CLI - "--intake <intake_json_path>" → from a saved/persisted intake - "--persist" → persist resulting packet to audit-logs/rtch/packets/ - "--handoff-to-consolidate" → after packet emission, hand selected_surfaces to /consolidate secondary_modulation_axes: - target_profile: doctrine_chain | audit_history | code_path | manifest_registry | vague_intent | mixed - fanout_level: conservative | normal | wide - mutation_risk: read_only | low_mutation | high_mutation - expected_output: hydration_packet | target_set_for_consolidate | single_chunk | claim_evidence IMPLEMENTATION_STATUS: binder: audit-logs/governance/runtime-tactical-context-hydration-binder.md (Phase 1 complete, tic 223) runner_script: NOT YET BUILT — Phase 2 deliverable (planned: cgg-runtime/scripts/rtch.py) current_mode: manual-discipline — agent walks the 8 stages using Read/Bash/Grep tools directly promotion_status: design lane, not doctrine; Phase 7 routes the doctrine question after Phase 6 validation
tools3
complement
Closure inference and response-geometry disclosure — direction-agnostic, scope-aware detection of materially missing expressions around an active move. CENTROID: closure inference at the point of an active move IS: - post-landing closure check (detect non-local incompleteness after a move lands) - origin-shape detection (dual-ray structure at response formulation, before commit) - surface/defer/suppress decision routing via the structural-relevance gate IS NOT: collapse_zones: - follow-up helper (complement does not generate next-steps unprompted) - pattern mining lite (complement is local coherence, not statistical surface) - documentation helper (complement does not restructure existing content) - decorative expander (must willingly say "current focus is sufficient" more often than surface) - autonomous widener (origin-shape yields shaping choice to the user; never commits the geometry) sibling_overlaps: - /review (constitutional judgment at the pattern level) - /siren (persistent-condition signal emission) - pattern mining (statistical surface discovery) WHEN: - after a local closure event that may hide non-local incompleteness - at the point of formulating a response with apparent dual-ray structure - when an artifact lands and the caller suspects it is partial - on explicit Architect invocation NOT WHEN: - after every trivial step (gates are narrow by design) - when the move is within a bounded scope already (compile fix, single-file edit) - when current focus must be protected (Mode C suppression is the most important mode) - when the candidate complement is decorative (no change to implementation, governance, proof, boundary, or sequencing) RELATES TO: - /review (constitutional judgment — review promotes lattice-gap findings; complement surfaces closure-geometry gaps) - /siren (signal emission — siren records persistent conditions; complement surfaces missing expressions in the active move) - pattern mining (statistical — pattern mining scans populations; complement operates on a single active move) ARGS: stance: lens off_envelope: proceed-with-note # off_envelope rationale: /complement has an established default (scan current # move, conversation-only context). Undeclared-arg most commonly means # "apply the default" rather than caller confusion — proceed-with-note # preserves the default while logging the invocation for calibration. core_dispatch_rays: - "" → post-landing (scan current move) - "--origin" → origin-shape (detect dual-ray before response commits) - "--governance" → post-landing with governance context (signals + CogPRs) - "--full" → post-landing with full context (+ pattern-mining briefing) secondary_modulation_axes: - emphasis: surface | defer | suppress - context_depth: conversation-only | governance | full
tools3