Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Orient at the start of a coding session, review what recently changed in a repository, and self-audit after completing work. Use when the user wants the daily briefing (what changed in the last 24h with complexity deltas), hotspots (complexity × churn refactor priorities), or a session review (clean/review/risky verdicts per editing session). For catching up after time away or resuming a prior session, use memtrace-session-continuity; daily is the last-24h briefing + hotspots + self-audit. Do no
.claude/skills/syncable-dev-memtrace-daily/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 136% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -32% | 0% |
Session bookends: orient before you start, audit before you finish. All three tools read the bi-temporal version history — every save is a change event with complexity deltas, so "what happened" includes "did it make the code better or worse".
| Tool | Purpose | |------|---------| | get_daily_briefing | Last 24h diffed at graph level: changed functions with complexity deltas, new functions, new endpoints, per-module distribution | | find_hotspots | Functions ranked by complexity × recent changes — the ordered refactor priority list | | review_agent_sessions | Editing sessions clustered by actor, judged clean / review / risky |
> Parameter types: MCP parameters are strictly typed. Numbers
Full parameter spec for every Memtrace tool: references/mcp-parameters.md (bundled at the memtrace-skills plugin root). > (window_hours, window_days, top_n) must be JSON numbers.
get_daily_briefing with repo_id. Read the summary first:
net_cyclomatic_delta positive → recent work added complexity; checkwhether your task touches the same area before piling on.
changed rows with large positive deltas → recently-destabilised code;prefer preflight_check before editing anything in that list.
find_hotspots — work the list top-down. A hotspot you simplify pays off on every future change; a complex-but-untouched function can wait.
review_agent_sessions and find YOUR session (newest, your agent id):
clean — done; state the verdict in your summary.review — re-read your top_changes rows; justify each complexityincrease or simplify it.
risky — do not hand off yet. You added >30 net cyclomatic or toucheda >50-complexity function; extract helpers / flatten nesting first, then re-run the review.
Net complexity delta ≤ 0 unless the task genuinely required new branching (new feature with new cases). "I left the code simpler than I found it" is verifiable here — verify it.
| Tool | Returns | |------|---------| | get_daily_briefing | summary with net_cyclomatic_delta; changed rows (function + complexity delta); new functions, new endpoints, per-module distribution | | find_hotspots | ranked list of functions scored by complexity × recent changes | | review_agent_sessions | per-session verdict (clean / review / risky) with top_changes rows and net cyclomatic delta |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 6,878 | 2,993 | -56% | 1 | 1 | 0% | 884 | 1,016 | +15% | 0 | 0 | — |
case-02 | fail→fail | 11,665 | 5,648 | -52% | 1 | 1 | 0% | 1,755 | 934 | -47% | 0 | 0 | — |
case-03 | fail→fail | 7,290 | 5,124 | -30% | 1 | 1 | 0% | 842 | 894 | +6% | 0 | 0 | — |
case-04 | pass→pass | 14,455 | 8,972 | -38% | 1 | 1 | 0% | 2,482 | 2,080 | -16% | 0 | 0 | — |
case-05 | pass→pass | 16,270 | 12,602 | -23% | 1 | 1 | 0% | 2,382 | 2,732 | +15% | 0 | 0 | — |
case-06 | pass→pass | 7,455 | 10,101 | +35% | 1 | 1 | 0% | 1,094 | 2,721 | +149% | 0 | 0 | — |
case-07 | pass→pass | 8,213 | 2,717 | -67% | 1 | 1 | 0% | 1,392 | 1,128 | -19% | 0 | 0 | — |
case-08 | fail→fail | 10,864 | 4,330 | -60% | 1 | 1 | 0% | 1,638 | 1,376 | -16% | 0 | 0 | — |
case-09 | fail→pass | 7,292 | 3,599 | -51% | 1 | 1 | 0% | 1,212 | 972 | -20% | 0 | 0 | — |
case-10 | pass→pass | 8,015 | 3,709 | -54% | 1 | 1 | 0% | 1,175 | 1,220 | +4% | 0 | 0 | — |
case-16 | fail→pass | 16,029 | 5,448 | -66% | 1 | 1 | 0% | 628 | 1,480 | +136% | 0 | 0 | — |
case-11 | fail→pass | 7,333 | 4,604 | -37% | 1 | 1 | 0% | 1,304 | 1,248 | -4% | 0 | 0 | — |
case-12 | pass→pass | 9,121 | 4,831 | -47% | 1 | 1 | 0% | 1,367 | 1,397 | +2% | 0 | 0 | — |
case-13 | fail→pass | 12,306 | 6,072 | -51% | 1 | 1 | 0% | 1,845 | 1,566 | -15% | 0 | 0 | — |
case-14 | pass→pass | 12,006 | 4,908 | -59% | 1 | 1 | 0% | 1,715 | 1,468 | -14% | 0 | 0 | — |
case-15 | fail→pass | 13,587 | 4,686 | -66% | 1 | 1 | 0% | 2,066 | 1,407 | -32% | 0 | 0 | — |
case-17 | fail→pass | 11,378 | 5,784 | -49% | 1 | 1 | 0% | 1,673 | 1,518 | -9% | 0 | 0 | — |
case-18 | pass→pass | 4,333 | 3,639 | -16% | 1 | 1 | 0% | 599 | 1,193 | +99% | 0 | 0 | — |
case-19 | fail→fail | 7,312 | 3,000 | -59% | 1 | 1 | 0% | 1,050 | 1,089 | +4% | 0 | 0 | — |
case-20 | pass→pass | 11,917 | 3,929 | -67% | 1 | 1 | 0% | 1,921 | 1,244 | -35% | 0 | 0 | — |
case-21 | fail→pass | 15,703 | 4,207 | -73% | 1 | 1 | 0% | 2,287 | 1,289 | -44% | 0 | 0 | — |
case-22 | fail→pass | 6,936 | 5,082 | -27% | 1 | 1 | 0% | 1,043 | 1,397 | +34% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 19 counted toward the lift figure. The other 3 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +36 percentage points is the difference between those two pass rates over the 19 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.