Content
88%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is a strong, execution-first skill document: concrete commands with correct flags, a numbered procedure with explicit validation steps, real pitfalls (the /exit trap, pty requirements), and a smoke-test verification section. The only weaknesses are mild — a slightly long inline keybindings/flags section and a brief narrative in Binary Resolution that could be tightened or moved to a reference file.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is efficient: nearly every line is a concrete command or a short, non-obvious operational fact ("Do NOT use /exit", "Enter may need to be pressed twice", "run does NOT need pty"). Minor over-explanation could be trimmed — the Binary Resolution narrative and parts of the TUI keybinding table (e.g. Ctrl+X session/editor shortcuts) are not strictly needed when driving via process actions — so it lands at anchor 4 rather than the fully lean anchor 5. | 4 / 5 |
Actionability | Every example is a complete, copy-paste-ready invocation with all relevant flags (workdir, background, pty, -f, --model), and coverage spans the common cases: one-shot run, interactive background session, resume, flags table, PR review, parallel worktrees, and a verification smoke test. This matches anchor 5 fully. | 5 / 5 |
Workflow Clarity | The Procedure section is a clearly numbered sequence that begins with an explicit validation checkpoint (verify version + auth list), includes monitoring (poll/log), input handling (submit), and clean exit, and the separate Verification section defines a smoke test with concrete success criteria ("Output includes OPENCODE_SMOKE_OK"). Error-recovery guidance ("If OpenCode appears stuck, inspect logs before killing") provides the feedback loop, matching anchor 5. | 5 / 5 |
Progressive Disclosure | No bundle files exist, and the ~200-line body is well-sectioned with clear headers (When to Use, Prerequisites, One-Shot Tasks, Interactive Sessions, Pitfalls, Verification). It is not a 5: the skill exceeds 50 lines, and reference-shaped material (the TUI keybindings table and the common-flags table) is inlined where a references/ file would keep SKILL.md leaner — a minor organization gap consistent with anchor 4. | 4 / 5 |
Total | 18 / 20 Passed |