CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

gap

Thin Codex skill shim for node_gap_compute. Use when running OmniNode integration gap detection, report classification, or gap-cycle preflight.

Quality

78%

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

Fix and improve this skill with Tessl

tessl review fix ./src/omnimarket/adapters/codex/skills/gap/SKILL.md

The canonical home for this skill is gap in OmniNode-ai/omnimarket

SKILL.md
Quality
Evals
Security

Quality

Content

75%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.

A tight, highly actionable dispatch reference: executable command, complete argument/payload mapping, and explicit output and error handling, with no filler. The remaining gaps are minor — a concrete example payload for the default detect case, deduplicated contract metadata, and a recovery path when ok is false.

Suggestions

Add one copy-paste-ready example payload for the default detect case so the dispatch command is fully executable end-to-end.

Deduplicate the contract metadata: command name, command topic, and terminal topic appear in both the generated runtime-truth block and the Contract section — keep a single source and point to it.

Complete the failure loop for ok=false: state when to retry (e.g., after fixing payload validation errors surfaced by --compile-only) versus when to stop and report the error to the user.

DimensionReasoningScore

Conciseness

The body is dense and system-specific with no explanation of concepts Claude already knows, but the command name ("gap_compute"), command topic, and terminal topic each appear twice — once in the generated runtime-truth block and again in the Contract section — which is trimming-level duplication, matching anchor 4 rather than the every-token-earns-its-place level of 5.

4 / 5

Actionability

The dispatch section gives a fully specified executable command (env -u PYTHONPATH uv run python scripts/run_codex_runtime_request.py with exact flags), the payload field mapping, and explicit output handling (output_payloads[0], error.code/error.message). It falls short of anchor 5 because '--payload <json-payload>' is a placeholder with no copy-paste-ready example payload for the common detect case.

4 / 5

Workflow Clarity

The dispatch workflow (map user inputs to payload fields → run wrapper → interpret ok/output_payloads or surface error) is clear and unambiguous, and preflight validation exists via '--compile-only'. It is not 5 because the failure path is only 'surface error.code and error.message' with no recovery or retry loop.

4 / 5

Progressive Disclosure

No bundle files exist and the body is well organized into Arguments, Dispatch, Contract, and Output sections with all content appropriately inline for a thin shim, with no nested references. It is not 5 because the ~40-line generated runtime-truth metadata block is inlined under the Arguments heading (a structural quirk that also creates the Contract-section duplication).

4 / 5

Total

16

/

20

Passed

Description

73%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.

A focused, correctly structured description with an explicit and concrete 'Use when' trigger clause. Its main weakness is the what-clause: 'thin Codex skill shim' names the mechanism without stating what the skill does, and the trigger set omits the fix and reconcile subcommands the skill supports.

Suggestions

Replace 'Thin Codex skill shim for node_gap_compute' with a what-clause that states the action taken, e.g. 'Dispatches gap_compute requests to node_gap_compute and renders gap findings, classifications, and skipped probes.'

Add triggers for the uncovered subcommands so users asking to 'fix gaps', 'run a gap cycle', or 'reconcile a gap report' match the skill.

DimensionReasoningScore

Specificity

"Thin Codex skill shim for node_gap_compute" names the domain but the what-clause describes no concrete actions of its own; the concrete operations ("gap detection, report classification, or gap-cycle preflight") appear only in the when-clause, matching anchor 3 (domain plus limited concrete actions, not comprehensive) rather than 4, which expects the skill's actions themselves to be listed.

3 / 5

Completeness

Both parts are present: "Thin Codex skill shim for node_gap_compute" (what) and an explicit "Use when running..." clause with three concrete triggers (when). It is not 5 because the what is weak — "shim" never says what the skill actually does; it is not 3 because the when is fully explicit.

4 / 5

Trigger Term Quality

"OmniNode integration gap detection", "report classification", and "gap-cycle preflight" are natural phrases a user in this domain would say, but common variations like "fix gaps", "reconcile a gap report", and "gap report" (all backed by the skill's subcommands) are missing, so it fits anchor 4 rather than the comprehensive coverage of 5.

4 / 5

Distinctiveness Conflict Risk

Highly niche terms ("OmniNode", "node_gap_compute", "gap-cycle") give the description a clear trigger niche with minimal overlap risk against other skills, matching anchor 5.

5 / 5

Total

16

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 3 missing

Warning

Total

15

/

16

Passed

Repository

OmniNode-ai/omnimarket

Path

src/omnimarket/adapters/codex/skills/gap/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.