tracedecay

Prefer tracedecay tools for codebase exploration, graph queries, and memory recall.

Use when installing a TraceDecay checkout for live local development, refreshing the globally available binary without a release, or validating agents and the daemon against unmerged TraceDecay changes.

Use when installing a TraceDecay checkout for live local development, refreshing the globally available binary without a release, or validating agents and the daemon against unmerged TraceDecay changes.

Use when listing or reading agent-managed automation skills, viewing automation run artifacts, or inspecting skills owned by the standard Hermes user install without mutating them.

Use when you need LCM, session search, transcript search, raw past-session replay, scoped/time grep, summary-DAG drill-down, branch/worktree/commit history, workflow recovery, or compaction recovery; and when driving host LCM lifecycle preflight/compress/boundary/repair.

Use before writing a new function or helper (it may already exist), and before any rename, signature, or field change — the graph lists every call site.

Use the moment the next action is Grep, Glob, rg, find, cat, or opening a source file to answer "where is X", "how does Y work", a type/trait question, or to check another git branch. Returns the enclosing symbol. Do NOT use for callers/callees (tracedecay:tracing-functions).

Use when unexpected changes appear on your branch or worktree — commits you did not make, amended or force-pushed history, or a working tree that contradicts your memory — to reconstruct who did what and decide how to proceed.

Use when Review self-improvement automation runs and activation policy.

Use when Review scheduler decisions before enabling automation.

Use when MCP is intentionally absent or unavailable, a host or subagent has shell access without MCP, or a tracedecay MCP call fails, times out, or disconnects.

Use when asking what TraceDecay can do, which TraceDecay tool, skill, command, or workflow fits a task, or how to discover capabilities when MCP is absent or optional.

Use when checking TraceDecay adoption or telemetry health — usage analytics, silent hooks, stale hook logs, or empty analytics tables. Uses the analytics, doctor, and sessions CLI, never raw database queries.

Use when estimating blast radius, finding what depends on a symbol or file, choosing or running affected tests, or verifying a change without a full suite. Trigger before guessing tests or running "cargo test" broadly. Do NOT use to review a diff (tracedecay:reviewing-changes).

Use when starting any code or repo-context task in a TraceDecay-indexed project, before Grep/Glob/cat/Read/gh pr diff/tests/memory writes, including cross-project or sibling workspace context via project registry. Skip only for scoped subagents handed exact files.

TraceDecay Dev: Use when auditing, testing, debugging, or improving TraceDecay hook hint behavior, especially Codex/Claude/Cursor hint injection, verbosity, repetition, dedupe, token efficiency, real transcript examples, synthetic scenarios, or `src/hooks/tool_hints*` changes.

TraceDecay Dev: Use for a full introspection-and-improvement pass over TraceDecay itself — triggers every dev skill in dependency order (diagnostics gate, evidence sweep across usage/logs/automation/hints/code-health, synthesis, fixes, verification). Use when asked to 'improve tracedecay from its own data', run a mass introspection, or when several dev skills would otherwise be invoked ad hoc.

TraceDecay Dev: Use when auditing, testing, debugging, or improving TraceDecay hook hint behavior, especially Codex/Claude/Cursor hint injection, verbosity, repetition, dedupe, token efficiency, real transcript examples, synthetic scenarios, or `src/hooks/tool_hints*` changes.

Use when reviewing a PR, branch, or working-tree diff, auditing ship risk (panic/unsafe/todo, dead code, untested hotspots), or drafting commit/PR/changelog text. Trigger before "gh pr diff" — pr_context/diff_context compute changed symbols offline. Do NOT use for pure test selection (tracedecay:assessing-impact).

Use when producing code-health scorecards, architecture maps, project identity/index status, registered project lookup, cross-project or cross-repo context gathering, TODO/config/runtime checks, or before/after refactor health deltas.

Use when about to save or recall anything durable: before writing MEMORY.md, auto-memory, or CLAUDE.md, before answering from a stale summary, or before web-searching what a prior session already answered. Covers store, recall, curation. Do NOT use for raw transcript replay. Never stores secrets.

TraceDecay Dev: Use when interpreting TraceDecay compiler diagnostics, `tracedecay tool diagnose`, `tracedecay tool diagnostics`, mapped errors, affected symbols, or build/type failure output.

TraceDecay Dev: Use when turning TraceDecay's own analytics, session history, automation runs, managed skills, memory facts, diagnostics, or code-health signals into repo improvements, evals, or bundled skill updates.

TraceDecay Dev: Use when mining TraceDecay session logs, analytics, diagnostics, automation artifacts, or agent transcripts to improve TraceDecay code, tools, or packaged skills.

TraceDecay Dev: Use when interpreting TraceDecay compiler diagnostics, `tracedecay tool diagnose`, `tracedecay tool diagnostics`, mapped errors, affected symbols, or build/type failure output.

TraceDecay Dev: Use when turning TraceDecay's own analytics, session history, automation runs, managed skills, memory facts, diagnostics, or code-health signals into repo improvements, evals, or bundled skill updates.

TraceDecay Dev: Use when mining TraceDecay session logs, analytics, diagnostics, automation artifacts, or agent transcripts to improve TraceDecay code, tools, or packaged skills.

TraceDecay Dev: Use when auditing TraceDecay automation loops, skipped runs, memory-curator/session-reflector/skill-writer output, apply policy, or run artifacts.

TraceDecay Dev: Use when creating, revising, validating, approving, installing, or auditing TraceDecay agent-managed automation skills and skill-writer drafts.

TraceDecay Dev: Use when auditing TraceDecay automation loops, skipped runs, memory-curator/session-reflector/skill-writer output, apply policy, or run artifacts.

TraceDecay Dev: Use when creating, revising, validating, approving, installing, or auditing TraceDecay agent-managed automation skills and skill-writer drafts.

Use when compiler or type errors are in play: pasted "cargo check"/clippy/tsc/pyright output, a failed build, or a planned type check. Trigger before running those in the shell — tracedecay maps diagnostics to the enclosing symbol with callers. Do NOT use for test failures.

Use when the task is call-relationship shaped: "who calls X", "what does X call", "trace this", references before a rename, shortest path, recursion, or hubs. Trigger before grepping a function name — grep misses dynamic dispatch. Do NOT use to locate a symbol (tracedecay:exploring-code).