Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when you need to install the embedded robot agents into .github/agents, .claude/agents, .cursor/agents, or .codex/agents, selecting the destination interactively and copying the embedded agent definitions from project assets. This should trigger for requests such as Install embedded agents; Bootstrap .github/agents; Bootstrap .cursor/agents; Bootstrap .claude/agents; Bootstrap .codex/agents; Copy robot agents. Part of Plinth Toolkit
.claude/skills/jabrena-005-agents-installation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-21 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -68% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 6% | 0% |
Install a predefined set of embedded agent definitions from repository assets into a user-selected target directory. This is an interactive skill.
What is covered in this Skill?
.github/agents, .claude/agents, .cursor/agents, or .codex/agents)assets/agentsThis skill installs only the embedded robot agents bundle and must ask for destination before writing files.
.github/agents, .claude/agents, .cursor/agents, or .codex/agents before installingreferences/005-agents-installation.mdAsk exactly one question to choose .github/agents, .claude/agents, .cursor/agents, or .codex/agents and wait for an explicit answer before copying files.
Step constraints:
Create the destination directory if needed, then copy all embedded agent files defined in the reference content, preserving filenames and warning before overwriting existing files.
Return a concise checklist with selected destination, created/updated files, overwrite actions, and an optional verification command.
For detailed guidance, examples, and constraints, see references/005-agents-installation.md.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→pass | 5,217 | 1,819 | -65% | 1 | 1 | 0% | 831 | 687 | -17% | 0 | 0 | — |
case-21 | fail→pass | 10,181 | 2,856 | -72% | 1 | 1 | 0% | 1,539 | 856 | -44% | 0 | 0 | — |
case-01 | fail→pass | 5,685 | 2,064 | -64% | 1 | 1 | 0% | 750 | 721 | -4% | 0 | 0 | — |
case-03 | fail→pass | 13,058 | 1,499 | -89% | 1 | 1 | 0% | 1,971 | 640 | -68% | 0 | 0 | — |
case-04 | pass→fail | 17,353 | 15,612 | -10% | 1 | 1 | 0% | 3,032 | 2,749 | -9% | 0 | 0 | — |
case-05 | pass→pass | 8,551 | 6,857 | -20% | 1 | 1 | 0% | 1,570 | 1,711 | +9% | 0 | 0 | — |
case-06 | pass→pass | 9,061 | 5,084 | -44% | 1 | 1 | 0% | 1,682 | 1,352 | -20% | 0 | 0 | — |
case-07 | fail→pass | 4,907 | 2,294 | -53% | 1 | 1 | 0% | 729 | 772 | +6% | 0 | 0 | — |
case-08 | fail→fail | 6,381 | 5,859 | -8% | 1 | 1 | 0% | 914 | 1,386 | +52% | 0 | 0 | — |
case-09 | fail→pass | 16,158 | 2,486 | -85% | 1 | 1 | 0% | 2,450 | 796 | -68% | 0 | 0 | — |
case-10 | fail→fail | 5,409 | 4,531 | -16% | 1 | 1 | 0% | 862 | 590 | -32% | 0 | 0 | — |
case-11 | fail→pass | 9,438 | 11,182 | +18% | 1 | 1 | 0% | 1,533 | 2,184 | +42% | 0 | 0 | — |
case-12 | fail→pass | 6,050 | 7,563 | +25% | 1 | 1 | 0% | 830 | 1,595 | +92% | 0 | 0 | — |
case-13 | pass→pass | 6,903 | 7,949 | +15% | 1 | 1 | 0% | 1,063 | 1,726 | +62% | 0 | 0 | — |
case-14 | fail→pass | 4,181 | 3,087 | -26% | 1 | 1 | 0% | 655 | 882 | +35% | 0 | 0 | — |
case-15 | fail→pass | 8,354 | 3,091 | -63% | 1 | 1 | 0% | 1,346 | 968 | -28% | 0 | 0 | — |
case-16 | fail→pass | 2,418 | 2,889 | +19% | 1 | 1 | 0% | 295 | 856 | +190% | 0 | 0 | — |
case-17 | fail→pass | 4,680 | 1,434 | -69% | 1 | 1 | 0% | 669 | 600 | -10% | 0 | 0 | — |
case-18 | pass→pass | 6,078 | 9,271 | +53% | 1 | 1 | 0% | 884 | 1,500 | +70% | 0 | 0 | — |
case-19 | fail→fail | 3,936 | 6,763 | +72% | 1 | 1 | 0% | 525 | 1,501 | +186% | 0 | 0 | — |
case-20 | pass→pass | 9,759 | 4,246 | -56% | 1 | 1 | 0% | 1,528 | 1,076 | -30% | 0 | 0 | — |
case-22 | pass→fail | 6,441 | 2,741 | -57% | 1 | 1 | 0% | 1,049 | 858 | -18% | 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 21 counted toward the lift figure. The other 1 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 +14 percentage points is the difference between those two pass rates over the 21 comparable cases. 3 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.