philotic-stack

Use this skill when reviewing, provisioning, or documenting MCP membrane tools for external clients such as Perplexity, Claude, or Codex. It keeps Muninn capture, LifeGraph tools, and raw graph/runtime tools separated.

Use this skill when authoring or revising an agent role lens before governed execution. It gathers the missing role definition inputs, explains the required payload shape, and prepares the payload for the role.create_or_update workflow.

Use at session start and closeout to retrieve and write durable Muninn continuity context.

Use when the user needs planning, decomposition, orchestration, or seam selection guidance.

Use when the user needs validation strategy, test selection, or confidence reporting.

Use this skill when an agent summarizes the operator's LifeGraph, answers "what is in my LifeGraph?", or turns LifeGraph records into planning context. It requires provenance-aware summaries that separate confirmed graph facts from seeded placeholders, inferred intent, and recommended next structure.

Create, inspect, enable, disable, and remove scheduled cron jobs on the hotel. Use when the operator or an autonomous workflow needs recurring work at a fixed schedule.

Use this skill when a client should adopt the shared Muninn memory workflow for continuity across sessions. Trigger when setting up Muninn-backed recall/write-back habits, wiring a client to the shared helper, or standardizing memory behavior across multiple cognitive clients.

Protocol for an agent to send a proactive, unsolicited message to a user or peer agent without waiting for a human trigger. Use when a cron job fires, a threshold is crossed, or a monitored condition is met.

Restore session continuity at the start of a new conversation by pulling current state from the hotel graph, memory, and role list — then synthesizing a working mental model before acting.

How an agent understands its own toolset profile, grows its capability set over time, and requests or registers new skills for its roles. The self-directed capability growth loop.

Diagnose and recover from a stuck, failed, or confused session state. Classifies the failure, chooses the minimal recovery path, and resumes work without operator intervention when possible.

Use this skill when you want to capture structured event data into a queryable local table — e.g. logging model routing signals, transcription events, or any recurring envelope your agent-graph neighborhood should track across sessions.

Use this skill when a specialist role needs a tool or capability it does not currently have. Hands back to the orchestrator with a structured request so the orchestrator can register and assign the skill, then return the role to continue its work.

Reference skill definition for Hermes to monitor the local iMessage database and forward messages from specified automated senders. Intended to be registered by Hermes' orchestrator via skill.register.

Use this skill when an agent wants to codify a recurring delegation pattern as a named, reusable skill and assign it to their own role.

Use this skill when deciding how deeply to validate a change or when explaining confidence in a code change. Trigger for test planning, selecting the right validation rung, or classifying confidence as test-green, smoke-green, or watched-live-green. Do not use this skill for runtime root-cause investigation or Philotic slice close-out unless validation choice is the main task.

Use this skill to run a Philotic retrospective after a meaningful slice, live-debug session, or end-of-day checkpoint. It structures the retro around what went well, what did not, how to do better, how SVE should change, and how to optimize SVE. Trigger when the user asks for a retrospective, retro, lessons learned, process reflection, or engine/process improvement.

Use this skill when debugging live runtime behavior in a multi-process or distributed system, especially IPC framing, socket registration, supervised process startup, routing/materialization, liveness, or request/response mismatches. Trigger when the user asks to watch a live stack, debug runtime integration issues, inspect process/socket state, or compare test behavior against real execution. This skill is general and should not assume a Rust-only stack.

Use this skill when source changes must be proven in an installed, supervised, or operator-facing runtime before claiming smoke-green or watched-live-green. Trigger for Homebrew/launchd rollouts, stale-binary suspicion, installed-vs-source drift, or live watch validation after deploy.

HIGH PRIORITY. Use this skill when changing Philotic architecture or workflow docs so domain, frontmatter, status truth, proposal links, and task links stay aligned automatically. Trigger for edits in docs/architecture/, docs/README.md, or docs/task.md when the work changes current truth, near-future seams, proposal status, or documentation structure.

Use this skill when designing or debugging how services, runners, workers, or components are materialized, become ready, go dormant, wake, or are placed in specific environments. Trigger for lazy startup, readiness policy, environment placement, runner lifecycle, or materialization fallback design. Do not use this skill for generic live-stack triage unless lifecycle and placement policy are the main seam.

Use this skill when breaking a larger task into bounded subagent work. Trigger when the work would benefit from parallel investigation, focused implementation, test generation, or runtime triage, and you need to decide what context to pass, what authority to retain, and how to avoid overlap or confusion.