CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

adversarial-pipeline

Thin Codex skill shim for node_adversarial_pipeline_orchestrator. Use when running the adversarial plan-to-ticket pipeline through runtime truth.

Quality

72%

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/adversarial-pipeline/SKILL.md

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

SKILL.md
Quality
Evals
Security

Quality

Content

67%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 a strong, actionable dispatch guide with concrete commands, argument semantics, and thoughtful failure handling (compile-only preflight, dry_run, explicit stop states). Its main weaknesses are redundancy between the generated runtime-truth block and the Contract section, and the absence of a worked example payload that would make the happy path fully copy-paste ready.

Suggestions

Deduplicate the Contract section (or the runtime-truth block) so the command name and topic pair appear once, trimming the most obvious repeated tokens.

Add one concrete example --payload JSON (e.g., topic plus min_findings_gate) so the primary dispatch path is copy-paste ready.

State the recommended order explicitly (compile-only preflight first, then live dispatch) to turn the existing validation options into a sequenced procedure.

DimensionReasoningScore

Conciseness

The body is dense and skill-specific with no padding of concepts Claude already knows, but the Contract section repeats the command name and both topics already stated in the generated runtime-truth block (the command name appears three times), and "Route: generic Pattern-B adapter transport" is unexplained jargon. This matches anchor 3 (mostly efficient but could be tightened); it is below anchor 4 because the duplication is more than minor trimming.

3 / 5

Actionability

The dispatch section gives an executable command ("env -u PYTHONPATH uv run python scripts/run_codex_runtime_request.py --command-name ..."), an argument table with defaults, an explicit payload field mapping, and clear ok/error output handling. It falls short of anchor 5 only because "--payload '<json-payload>'" is a placeholder with no worked example payload, leaving the most common execution path not quite copy-paste ready.

4 / 5

Workflow Clarity

The sequence is clear (run location, map arguments to payload, dispatch via wrapper, branch on ok/error, render outputs) with validation checkpoints present: "--compile-only" preflight, "dry_run" for ticket creation, and the explicit "node_not_implemented" stop state. This matches anchor 4 (clear sequence, most checkpoints); it is not 5 because there is no explicit feedback loop or ordered preflight-then-dispatch procedure.

4 / 5

Progressive Disclosure

No bundle files exist, and all referenced paths (scripts/run_codex_runtime_request.py, the node directory) are clearly-labeled external repo paths rather than dangling references. Sections are well organized (Arguments, Dispatch, Contract, Output); the only organization gap is the ~40-line generated contract block duplicated in the Contract section. This is anchor 4 (good structure, minor gaps); not anchor 5 since the body is over 50 lines with contract metadata that could be consolidated.

4 / 5

Total

15

/

20

Passed

Description

66%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 highly distinctive and has an explicit 'Use when' trigger, but it leans on internal jargon and identifies the backing node rather than stating the concrete capabilities the pipeline provides. Trigger-term coverage is narrow, relying on one domain phrase plus technical identifiers users are less likely to say verbatim.

Suggestions

State the concrete capabilities in the description itself (e.g., 'runs design-to-plan, hostile-review gating, and Linear ticket creation') instead of only naming the backing node.

Add natural trigger variations users might say, such as 'adversarial pipeline', 'hostile review gate', or 'create tickets from a design topic'.

Drop or soften pure jargon like 'runtime truth' in the description, or pair it with a plain-language gloss so the trigger phrase reads naturally.

DimensionReasoningScore

Specificity

"Thin Codex skill shim for node_adversarial_pipeline_orchestrator" names a concrete backing node and "running the adversarial plan-to-ticket pipeline" is one concrete action, but the description never enumerates what the pipeline actually does (design-to-plan, hostile-review gating, ticket creation). This matches anchor 3 (domain plus 1-2 concrete actions, not comprehensive); it is above anchor 2 because it is not merely generic, and below anchor 4 because it lists no specific action set.

3 / 5

Completeness

An explicit "Use when running the adversarial plan-to-ticket pipeline through runtime truth" clause answers 'when', and the shim-for-node framing answers 'what' structurally. Both are present (anchor 4) but not anchor 5: the 'what' describes the artifact's shape rather than its function, and there are no concrete trigger phrases beyond the single 'Use when' clause.

4 / 5

Trigger Term Quality

"adversarial plan-to-ticket pipeline" is the natural in-domain phrase a user would say, but it is wrapped in heavy jargon ("Codex skill shim", "runtime truth", the raw node id) with no synonyms or variations such as "hostile review", "design review", or "Linear tickets". Some relevant keywords exist, but common variations are missing, matching anchor 3 rather than anchor 4's good coverage.

3 / 5

Distinctiveness Conflict Risk

The description names a unique node id and a unique pipeline name, giving it a clear niche with distinct triggers and minimal conflict risk with any other skill. It cleanly matches anchor 5; no neighboring anchor fits better.

5 / 5

Total

15

/

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/adversarial-pipeline/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.