CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

grilling-for-routine

Use when ask-dora routes a routine idea, when running dora harness new, or when grilling a loop-able idea into a routine. Collect skills to run, skills to refer to, the Scalekit Agent Gateway MCP URL, and a Hermes model, provider, and reasoning level from `dora harness models`. After one good one-pass, may freeze a deterministic step as a Fixed step.

66

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

71%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is exceptionally actionable — exact commands, paths, URLs, and flags for every operation — with a clear ordered workflow and real validation gates. Its weaknesses are structural: a ~47-line MUST recap that duplicates the sections, a cluttered step 10, and no offloading of detailed material (routine-copy provenance rules, post-save login operations) into reference files.

Suggestions

Delete or drastically prune the trailing MUST block — it restates rules already covered in the Loop-able, Pocket, Fixed step, Gate, and Routine copy sections, cutting roughly 45 lines of duplication.

Split the detailed operational rules into a one-level-deep reference file (e.g., references/routine-copy.md for provenance/origin rules and references/harness-ops.md for the step-10 login/apply/post-save commands), keeping SKILL.md as a lean overview plus the Order sequence.

Restructure the post-save operations currently jammed into Order step 10 into their own flat 'After save' section with a consistent sub-list, so login, apply, show/logs/rm, and model-change guidance are not nested under one numbered step.

DimensionReasoningScore

Conciseness

The body is telegraphic with zero concept-explanation filler, but the final ~47-line MUST block (lines 178-224) almost entirely restates rules already stated in the sections ('Do not invent a catalog' appears three times, 'The teammate can skip' twice). Mostly efficient, but the duplicate recap section could be cut.

3 / 5

Actionability

Fully executable, copy-paste-ready commands throughout: `hermes config get delegation.subagent_auto_approve`, `hermes mcp add scalekit --url <mcp-url> --auth oauth`, the complete `dora harness new --accept --yes …` invocation with every flag, and the official install URL.

5 / 5

Workflow Clarity

The 'Order' section gives a numbered 10-step sequence with explicit validation gates (loop-able check, pocket check, one-pass before freeze, spec check, gate before save, 'Do not fake a pass') and a retry loop ('Another one-pass is allowed'). Step 10 is a sprawling nested blob mixing apply/boot/login/post-save operations with inconsistent indentation, holding it below 5.

4 / 5

Progressive Disclosure

No bundle files exist and all ~225 lines of dense operational detail live inline in SKILL.md with clear section headers — well-navigable, but this is the full operational manual rather than an overview pointing to one-level-deep reference files. Matches the '200 lines that could be in a separate file' pattern with good headers.

3 / 5

Total

15

/

20

Passed

Description

87%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is strong: it explicitly states what the skill collects and when to invoke it, with concrete, ecosystem-specific triggers and exact command names. Minor gaps in action coverage (the loop-able/pocket checks) and a few missing natural synonyms keep specificity and trigger quality at 4 rather than 5.

DimensionReasoningScore

Specificity

Names several concrete actions — 'Collect skills to run, skills to refer to, the Scalekit Agent Gateway MCP URL, and a Hermes model, provider, and reasoning level from `dora harness models`' and 'may freeze a deterministic step as a Fixed step' — but the core loop-able/pocket gate checks are not surfaced, so coverage has minor gaps.

4 / 5

Completeness

Explicitly answers both: 'what' (collect the six gate items from `dora harness models`, optionally freeze a deterministic step as a Fixed step) and 'when' (three concrete 'Use when…' triggers).

5 / 5

Trigger Term Quality

'Use when ask-dora routes a routine idea, when running dora harness new, or when grilling a loop-able idea into a routine' gives good natural ecosystem phrases (routine idea, loop-able, grilling) plus a command trigger, but misses common variations like 'set up a routine' or 'night job'.

4 / 5

Distinctiveness Conflict Risk

Triggers are tightly scoped to ask-dora routing and `dora harness new` — a clear niche with distinct trigger phrases and minimal overlap risk against sibling skills.

5 / 5

Total

18

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

saif-shines/doraval

Path

skills/grilling-for-routine/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.