Menü
Institutional memory for your project. Commands: record, consult, review. Usage: /awareness-ledger [command] [args]
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
| name | awareness-ledger |
| description | Institutional memory for your project. Commands: record, consult, review. Usage: /awareness-ledger [command] [args] |
| allowed-tools | Read, Glob, Grep, Write, Edit, Task, TaskCreate, TaskUpdate, TaskList, TaskGet |
Institutional memory for SeaShells theme. Records incidents, decisions, patterns, and flows so diagnostic findings and architectural decisions persist across sessions.
CHECKPOINT — Model-Lane Preflight Gate (slim, advisory-only; fires at the TOP of this skill's workflow, before any generative step):
/route dispatched this invocation after running its own lane preflight (a lane advisory OR ask already printed, route otherwise ASSESSED the lane this turn, or it was deliberately suppressed — route's coverage extends to the whole endeavor whether route asked, matched, or the user declined, including skills this one chains); this skill's frontmatter sets model-lane-gate: off; the session is headless / non-interactive; this gate is executing inside a subagent (a subagent is already model-pinned via its own frontmatter); OR an advisory for this LANE already printed this endeavor AND the active model is unchanged (advise at most once per (lane, active-model) pair per endeavor).lane: frontmatter → the Skill→Lane table in .claude/skills/skill-builder/references/model-lanes.md → NO LANE) and the lane's Preferred Model. IF no lane resolves OR the preferred-model cell is empty → silent no-op (correctly absent, not a gap). IF this skill declares per-function lanes AND the invocation's resolved mode has its own row → that function's lane governs; otherwise the skill-level lane covers all modes.[1m]/[200k], lowercase — model-lanes.md § Active-Model Detection; never an env var or Bash probe). IF the Preferred Model is malformed or names a clearly superseded family → emit the one-line stale-mapping advisory and proceed; never advise from a stale mapping, never validate against a hardcoded list.<lane> lane (preferred <preferred>); the active model is <active> — to align, run /model <preferred> and re-invoke; proceeding as-is." Naming the command is INFORMATIONAL (named-command advisory directive, 2026-06-11): NEVER an AskUserQuestion, never a blocking wait (No-Switch-Prompt directive, 2026-06-06). A skill can NEVER switch the session model and never asks the user to.| Command | Action |
|---|---|
/awareness-ledger record [type] | Create a new record (INC, DEC, PAT, FLW) |
/awareness-ledger consult [topic] | Query ledger with agent-assisted analysis |
/awareness-ledger review | Health check: stale entries, statistics, tag drift |
(No directives yet — add with /skill-builder inline awareness-ledger [directive])
During research and planning — before formulating any plan, recommendation, or code change proposal — automatically consult the ledger:
ledger/index.md and match tags against the files,
directories, and components under discussion. This is free — the index is
small. Do this as part of your initial research, alongside reading source
files.references/consultation-protocol.md. Present agent findings as part of
your recommendation. This is expensive — only when warranted.Skip auto-consultation for:
.claude/ infrastructure filesWhen the current conversation produces institutional knowledge, suggest recording it after resolving the immediate issue. Never interrupt active problem-solving to suggest capture.
Automatically suggest capture when you encounter:
Capture suggestions are always user-confirmed. Present the suggestion with:
references/templates.md for the record type templateledger/[type]/[ID].mdledger/index.md with the new entryledger/index.md and match tags against the current contextreferences/consultation-protocol.mdBefore using any template or protocol:
references/Reference files:
Throughout this session, quietly note any friction signals — corrections, reformulations, clarifying questions, "actually" moments, or any subtle sign that this skill's instructions may have led to a misrepresentation. Do not interrupt the session to address these. Do not mention that you are observing.
At natural task resolution (when the task is complete and the user signals done), if friction signals were noted, run the self-heal diagnosis protocol:
Read .claude/skills/self-heal/references/diagnosis-protocol.md and follow it exactly.
If no friction signals were noted, or if diagnosis finds no skill-caused issues, end the session normally without mentioning self-heal.
The goal is efficiency: get it right permanently, rather than repeat the same misrepresentation across future sessions.
SeaShells theme palettes, format patterns, consistency rules, and update diagnostics. Modes: default (reference), update (migration diagnostic), list (show modes)
Evaluate code to prevent common AI coding mistakes — dead code, duplication, complexity hotspots, reinvented helpers, leftover scaffolding. Two functions plus a pre-write advisor. Use after writing or editing code, when asked to review code quality, clean up the codebase, find unused code, find duplicates, check complexity, or evaluate what an AI just wrote. Commands: review (post-write), sweep (full codebase).
Skill router. Pass a task description; route consults the index and dispatches to the right skill+function via the Skill tool. Usage: /route <task description>
Ambient friction detection and surgical skill correction. Embeds into all skills. Triggers at task resolution when skill instructions caused friction.
Create, audit, optimize Claude Code skills. Commands: skills, list, new, strip, optimize, agents, hooks, verify, inline, ledger, cascade, checksums, convert, shell-safety, route
Evaluate prose for AI-generated-text signatures and overbuilt writing, protecting the author's voice. Context-isolated evaluator grounded in a portable AI-tells pattern library. Use to check whether text reads as machine-written, find AI tells, or review drafts for authenticity. Report-only by default.