Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate a compact first-pass repository briefing with context-pack. Use when work starts in an unfamiliar repo, when the user asks for orientation or high-signal files, when active changes need a compact summary, or when prompt budget matters.
.claude/skills/hashgraph-online-context-pack/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 1095% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -54% | 0% |
Use this skill to turn a repository into a small, prioritized briefing before deeper exploration.
If the context-pack MCP server is installed through this plugin, prefer the MCP tools over shelling out manually:
get_contextget_changed_contextget_file_excerptinit_memoryrefresh_memorycontext-pack --cwd <repo> or the get_context MCP tool before a manual tree walk unless the task is already extremely narrow.context-pack --cwd <repo> --changed-only --no-treecontext-pack --cwd <repo> --format jsoncontext-pack --cwd <repo> --changed-only --no-tree --max-bytes 2000--include, --exclude, --max-files, or --max-depthcontext-pack --cwd <repo> --init-memorycontext-pack --cwd <repo> --refresh-memoryinit_memory / refresh_memory MCP toolsAGENTS.md, missing README, disabled git, or truncated output.context-pack command you used when it affects the result.--changed-only to keep the bundle focused on active files and diffs.--format json when another tool or script needs the result..context-pack/memory.md when repo-authored docs are weak.tree, blind rg, or random large-file reads for the first pass when context-pack is available.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-06 | fail→pass | 22,694 | 11,030 | -51% | 1 | 1 | 0% | 2,426 | 1,626 | -33% | 0 | 0 | — |
case-11 | pass→pass | 13,631 | 7,873 | -42% | 1 | 1 | 0% | 1,265 | 955 | -25% | 0 | 0 | — |
case-01 | fail→fail | 11,421 | 13,849 | +21% | 1 | 1 | 0% | 2,018 | 889 | -56% | 0 | 0 | — |
case-02 | fail→fail | 15,286 | 17,034 | +11% | 1 | 1 | 0% | 390 | 1,108 | +184% | 0 | 0 | — |
case-03 | fail→pass | 9,802 | 11,340 | +16% | 1 | 1 | 0% | 210 | 2,509 | +1095% | 0 | 0 | — |
case-04 | fail→pass | 17,807 | 9,271 | -48% | 1 | 1 | 0% | 1,784 | 1,375 | -23% | 0 | 0 | — |
case-05 | fail→pass | 15,551 | 9,820 | -37% | 1 | 1 | 0% | 1,372 | 1,354 | -1% | 0 | 0 | — |
case-07 | pass→pass | 16,159 | 7,191 | -55% | 1 | 1 | 0% | 1,785 | 1,610 | -10% | 0 | 0 | — |
case-08 | fail→pass | 20,609 | 2,474 | -88% | 1 | 1 | 0% | 2,232 | 1,020 | -54% | 0 | 0 | — |
case-09 | pass→pass | 16,404 | 7,037 | -57% | 1 | 1 | 0% | 1,675 | 1,469 | -12% | 0 | 0 | — |
case-10 | fail→pass | 16,994 | 11,953 | -30% | 1 | 1 | 0% | 2,031 | 1,650 | -19% | 0 | 0 | — |
case-12 | pass→pass | 22,039 | 11,436 | -48% | 1 | 1 | 0% | 2,229 | 1,573 | -29% | 0 | 0 | — |
case-13 | pass→pass | 16,179 | 7,419 | -54% | 1 | 1 | 0% | 1,579 | 1,625 | +3% | 0 | 0 | — |
case-14 | pass→pass | 12,626 | 7,431 | -41% | 1 | 1 | 0% | 1,007 | 991 | -2% | 0 | 0 | — |
case-15 | fail→pass | 18,530 | 2,772 | -85% | 1 | 1 | 0% | 1,935 | 913 | -53% | 0 | 0 | — |
case-16 | pass→pass | 18,851 | 6,771 | -64% | 1 | 1 | 0% | 2,212 | 1,529 | -31% | 0 | 0 | — |
case-17 | fail→pass | 5,755 | 2,027 | -65% | 1 | 1 | 0% | 906 | 905 | -0% | 0 | 0 | — |
case-18 | pass→pass | 21,103 | 13,603 | -36% | 1 | 1 | 0% | 2,135 | 1,927 | -10% | 0 | 0 | — |
case-19 | pass→pass | 12,042 | 9,942 | -17% | 1 | 1 | 0% | 1,889 | 1,470 | -22% | 0 | 0 | — |
case-20 | pass→fail | 8,199 | 13,173 | +61% | 1 | 1 | 0% | 1,244 | 1,175 | -6% | 0 | 0 | — |
case-21 | pass→fail | 18,554 | 4,776 | -74% | 1 | 1 | 0% | 2,812 | 829 | -71% | 0 | 0 | — |
case-22 | pass→fail | 9,925 | 15,533 | +57% | 1 | 1 | 0% | 2,060 | 924 | -55% | 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 16 counted toward the lift figure. The other 6 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 +23 percentage points is the difference between those two pass rates over the 16 comparable cases. 4 cases got worse with the skill loaded, and they are included in that figure.
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.