Content
92%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.
An excellent operational skill body: exact commands, a validated 3-step setup with explicit failure handling and a recovery-oriented troubleshooting table, and clear boundaries on what the automated agent sees and may do. Implementation detail lives in the bundled helper script rather than the body. The only trimmable content is a small amount of explanatory prose about event flow.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense with operational specifics and assumes competence (no explanation of what cron, daemons, or globs are), e.g. "AutoRAG's daemon sees files change and batches them (default debounce 2 s)". Minor instances of prose that could be trimmed — the 'How one event travels' walkthrough and the merging note — keep it at the 4 anchor rather than 5's 'every token earns its place'. | 4 / 5 |
Actionability | Fully executable, copy-paste-ready commands throughout: `python3 "${HERMES_SKILL_DIR}/scripts/autorag_automation.py" setup --id invoices --path ~/Downloads --include '**/*.pdf' --events create`, the npx daemon-install command, and list/remove invocations, plus exact cronjob field values and required prompt-ending text. This matches the 5 anchor's 'copy-paste ready... common cases covered', verified against the actual script's flags. | 5 / 5 |
Workflow Clarity | The setup is a clearly sequenced 3-step process (write rule/scripts, create cron job, start daemon) with explicit validation: "The helper validates the rule with AutoRAG's own parser before it changes anything" and "If it exits non-zero, show the user its `error` and stop; do not retry with guessed flags", plus a troubleshooting table with recovery paths (e.g. run fire.sh by hand, re-run setup after a Hermes update). The batch-operation cap does not apply because validation and feedback loops are present; this fits the 5 anchor, not the 4 anchor's 'minor validation gaps'. | 5 / 5 |
Progressive Disclosure | The body is a well-organized overview (setup, list/remove, limits, troubleshooting) with implementation detail correctly delegated to the one real bundle file, `scripts/autorag_automation.py`, referenced via `${HERMES_SKILL_DIR}` and confirmed to exist with matching subcommands. Nothing that belongs in a separate reference file is inlined, and navigation via section headers is easy — matching the 5 anchor for a skill of this scope. | 5 / 5 |
Total | 19 / 20 Passed |