Claude Skills by OutlineDriven

Use when pinning down domain terminology, maintaining the domain model, or when a term conflicts or needs sharpening. Resolves each term to one canonical owner, records it in CONTEXT.md with rejected synonyms, and keeps the glossary implementation-free.

Use when roadmap, plans, or docs may have drifted from code, or when restarting a stalled project. Returns an evidence-cited Reality Check Report with drift analysis and a prioritized reconstruction plan. Not for PR doc sync — use docs-update.

Use when a concept needs practising: the user asks for an exercise, quiz, recall, or gap probe. Runs one scaffolded practice move and records every attempt in PROGRESS.md. Not for explanations — use explain-concept.

Use when the user requests an engineering retrospective for a named period. Discovers repository and tracker telemetry, establishes source precedence, derives team breakdowns and evidence-backed habits, and writes a single report file. Not for an agent-environment retrospective — use agent-environment-retrospective; for a learning milestone — use learning-retrospective.

Use when reviewing an evaluation, benchmark, or scoring harness for leakage or contamination. Returns a read-only audit naming detected leakage patterns, where independent ground truth enters or fails to enter, and a fix for each finding.

Use when a concept needs making clear rather than practising: explain it simply, why does this exist, draw it, what is the difference, or simplify it for a beginner. Delivers a one-screen explanation from one chosen angle, grounded in sources. Not for scaffolded practice — use drill.

Use when asked to create a durable local explainer document for a concept, diff reference, idea, or work recap. Classifies the input, gathers grounding material, offers a quiz, drafts and structurally verifies the artifact in a scratch directory, and presents it. Not for one-screen explanations — use explain-concept.

Use when asked to cold-read an artifact with fresh zero-context eyes and cut whatever a stranger cannot follow. Dispatches blind sub-sessions that read only the inline artifact, compares their understanding against a private intent note, and reports severity-ordered defects. Not for criteria-review of requirements docs — use doc-review.

Use when a user wants an answer only from the breaking seat (stakeholders, dependencies, accounting, compatibility under destructive pressure). A breaking-perspective answer is emitted without blending. Not for any other from-* lens seat or multi-lens synthesis.

Use when the user wants an answer only from the business seat, to produce a business-perspective analysis of money, customers, and timing without blending other lenses. Not for any other from-* lens seat or multi-lens synthesis.

Use when a user wants an answer only from the career seat (effects on human trajectories). Emits a career-perspective analysis without blending other lenses. Not for any other from-* lens seat or multi-lens synthesis.

Use when asked to answer only from the codebase seat: what existing code tolerates or punishes. Emits a single-perspective analysis without blending. Not for rebuilding from primitives — use from-first-principle. Read-only; no source or remote mutation.

Use when a user wants an answer only from the human seat: what a person can love, trust, and tolerate. Emits a human-perspective analysis without blending. Not for rebuilding from primitives — use from-first-principle. Read-only; no source or remote mutation.

Use when asked to answer only from the impact seat: who and what actually moves. Returns an impact-perspective analysis without blending. Not for rebuilding from primitives — use from-first-principle. Read-only; no source or remote mutation.

Use when the user wants an answer only from the innovation seat: original technique, talent, and culture. Emits an innovation-perspective analysis without blending. Not for rebuilding from primitives — use from-first-principle. Read-only; no source or remote mutation.

Use when the user wants an answer only from the moat seat: building, keeping, and thickening defensibility. Emits a moat-perspective analysis without blending. Not for rebuilding from primitives — use from-first-principle. Read-only; no source or remote mutation.

Use when the user wants an answer only from the rent-seeking seat: extraction without building. Emits a rent-seeking-perspective analysis without blending. Not for rebuilding from primitives — use from-first-principle. Read-only; no source or remote mutation.

Use when a user wants an answer only from the skeptic seat: cold reasoning without project loyalty. Emits a skeptic-perspective analysis without blending. Not for rebuilding from primitives — use from-first-principle. Read-only; no source or remote mutation.

Use when the user wants an answer only from the stability seat: preservation of the working machine. Emits a stability-perspective analysis without blending. Not for rebuilding from primitives — use from-first-principle. Read-only; no source or remote mutation.

Use when asked to summarize feedback on the active PR. Returns severity-grouped feedback and an action list. Not for finding PRs to review — use gh-review-requests. Not for resolving feedback — use resolve-pr-feedback. Read-only.

Use when the user asks about recent engineering work or what the team is working on, or is preparing planning or roadmap material. Also handles an optional Slack summary when the user explicitly requests it. Not for remote mutation or any irreversible change.

Use when facing a bug, integration failure, dependency issue, unclear API usage, implementation blocker, or tool capability need and the user wants open-source solutions from GitHub. Searches issues, PRs, code, and repositories via `gh`, ranks by problem fit, and returns a reuse/adapt/avoid recommendation with a verification command. Not for source or remote-system changes.

Use when starting a codebase or service, scaffolding, migrating or refactoring, picking a language edition, runtime, framework, or dependency, or when the user asks for the latest, current, LTS, or modern way. Not for remote, credential, publish, deploy, or irreversible changes.

Use when a user resumes work and needs an orientation card without full human recall; returns a small card of shape, invariants, and smell locations. Not for packaging context for another agent or session — use handoff; never source or remote-system changes.

Use when substantive work reaches a session boundary, resumes after interruption, completes, or must move to another session, agent, harness, directory, or repository. Maintains ignored local continuity state and emits a bounded portable brief without launching its receiver.

Use when the user asks to open or take over a visible browser session, or needs to intervene in a live browser by hand, for example to solve a CAPTCHA or authenticate. Use for the headed takeover handshake: pre-flight cleanup, headed launch, mode confirmation, and a round-trip proof the user can see. Not for headless scraping or unattended automation.

Use when the user human-confirmedly asks to remove a session or note from recall. Writes a concrete exclusion or tombstone marker to a local overlay schema, atomically rebuilds the recall index, and supports unforget and list. Not for remote, credential, publish, deploy, or irreversible changes.

Use when a user asks to audit what recall fed agents. Returns a bounded event table of kind, time, session count, bytes, and empty-result flag, or a null object when absent, and refuses digest text replay. Returns an explicit error when the CLI is missing rather than conflating it with an empty log. Not for tasks that require source or remote-system changes.

Use when the user says remember this or settles one durable fact, append one bounded redacted note as strict UTF-8 JSONL with RFC3339 time and project to the local append-only store. Don''t use for remote, credential, publish, deploy, or irreversible changes.

Use when a user requests the reasoning behind one matched session by query or handle. Returns a capped record of decisions, outcome, touched files, lifecycle, and spawn edges. Not for peer transfer — use history-sync; not for store registration — use history-source-registry.

Use when a coding-agent session store is added or its format drifts, to document its layout, roles, time handling, and quirks in the registry with a passing conformance fixture. Not for recalling a session — use history-recall; not for peer transfer — use history-sync.

Use when the user requests memory transfer to or from a named peer. Exports redacted static JSONL batches, imports them idempotently, and advances per-peer watermarks. Not for recalling a session — use history-recall; not for store registration — use history-source-registry.

Use when the user asks "how does X work" and wants an architecture walkthrough of the code path or subsystem, including its gotchas. Don''t use for tasks that require source or remote-system changes.

Use when the user asks to park ideas or inspiration for later. Stores one short project card per qualifying item under idea-sparkbox/ without organizing it. Never stores code, backlog, or divergence-class cards. Don''t use for remote, credential, publish, deploy, or irreversible changes.

Use when an agreement-seeking interaction arises, including mid-conversation moments the model detects. Produces a prepared mandate with mutual-gains agenda, live response scripts, or a retrospective debrief with pathology analysis.

Use when a user, especially a new hire, asks for mentoring, guidance, or explanation of infrastructure or engineering practices. Returns a clear, sourced explanation with references and concrete next learning steps.

Use when a user asks to define requirements or choose between options. A structured interview surfaces every unstated assumption one question at a time and produces a user-approved decision record before any file change. Not for intent-from-data — use intent-proposal.

Use when a non-trivial change lands or is abandoned and the decision, its rejected alternatives, and the reason must be captured in a local topic file. Records Decision, Rejection, Reason, Type, Status, and Evidence with a confirmation gate before writing. Not for remote, credential, publish, deploy, or irreversible changes; not for trivial or corrective changes — those belong in CHANGELOG.

Use when departing knowledge must enter project topic files through a narration-first interview, targeted gap closure, privacy filtering, and Source = interview. Don''t use for remote, credential, publish, deploy, or irreversible changes.

Use when contradictions, revisit conditions, or duplicates appear in knowledge entries to resolve them by flipping statuses, marking superseded content, merging duplicates, or proposing file splits. Don''t use for remote, credential, publish, deploy, or irreversible changes.

Use when project documentation needs a layout or a knowledge item needs one home. Proposes a topic-indexed why-layer only where no suitable structure exists. Don''t use for remote, credential, publish, deploy, or irreversible changes.

Use when an existing repository needs its unexplained rationale recovered into topic files. Classifies evidence and status, marks unrecoverable items unknown, and leaves source conflicts open. Don''t use for remote, credential, publish, deploy, or irreversible changes.

Use when a project context-schema differs from the installed entry format. Classifies behind/current/ahead, consent-gates migration, marks missing values undefined, and advances only after every entry catches up. Don''t use for remote, publish, deploy, or irreversible changes.

Use when a knowledge artifact needs review before sharing or execution. Runs strategic and data reviewers in parallel, merges P1/P2/P3/Clean findings, and blocks ordinary shipping on P1. Don''t use for tasks that require source or remote-system changes.

Use when asked to review learning progress after a milestone. Produces a learning retrospective with an adjusted plan and a measurable next milestone. Not for an engineering retrospective — use engineering-retrospective; for an agent-environment retrospective — use agent-environment-retrospective.

Use when the user points to a folder of their study material and wants it made teachable. Produce a source inventory and a prerequisite-ordered concept map in CORPUS.md. Don''t use for remote, credential, publish, deploy, or irreversible changes.

Use when the user and agent may hold different pictures of current state after a gap, sync request, or exposed stale claim. Produces an evidence-backed ledger of agreed facts, corrected beliefs, and open assumptions. Not for persistence — use memory-update.

Use when the user asks for onboarding, orientation, a repository tour, or where to start. Produces an evidence-backed seven-section map, deep-reads real files, and asks for the next move.

Use when the user asks whether a repository is ready for public release or wants a gap assessment. Runs a read-only local audit of secrets, docs, licensing, CI, and packaging; performs optional remote checks only when a GitHub URL is supplied. Not for choosing or applying a license — use open-source-license-selection.

Use when asked to explain or teach a subsystem, module, pattern, or change in progressive layers from purpose to code depth. Grounds each layer in concrete file paths, line ranges, and symbols. Not for design rationale — use why. Not for source or remote-system changes.