round-verification

Verify integrated round evidence and return a machine-readable round result for script import.

Quality

68%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./docs/plantree/plans/agentic-loop-workflow/drafts/agentroles.ccb_round_reviewer/skills/round-verification/SKILL.md

Quality

Content

78%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 content is a tight, executable instruction-only skill: it specifies the exact output contract, an explicit gate checklist, and clear boundaries with no filler. Minor gains are available from deduplicating the tools/tests prohibition and giving one worked example of the result line.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes competence ('Reject missing node review, integration drift, ...' with no concept explanations), but there is mild redundancy: the no-tests/no-tools constraint appears both in the workflow prose ('Do not run tests, tools, shell commands...') and again in Boundaries ('Do not run tests or tools.'). That trims to anchor 4 ('minor instances of over-explanation that could be trimmed') rather than 5.	4 / 5
Actionability	The guidance is highly concrete: an exact output template 'round result: pass\|partial\|replan_required\|blocked', a hard rule that it must be the first non-empty line, and a specific fallback ('If the evidence is insufficient, start with round result: blocked'). It falls short of anchor 5 only because there is no worked example of a pass vs. blocked result line and the required evidence items in step 1 are listed without paths or formats.	4 / 5
Workflow Clarity	The three-step sequence (read evidence, gate on rejection conditions, emit result) is clear, and the rejection checklist plus the insufficient-evidence fallback act as explicit validation checkpoints — close to anchor 5. It stays at 4 because step 1 packs many evidence types into one long sentence with no indication of where each ref lives, and there is no explicit recovery loop if the first-line rule is accidentally violated.	4 / 5
Progressive Disclosure	The skill is under 50 lines, needs no external references (none exist in the bundle), and is well organized into a workflow, an output-format block, and a boundaries section. Per the scoring note for simple skills under 50 lines with no need for external references, this matches the top anchor.	5 / 5
	Total	17 / 20 Passed

Description

57%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 concise, in third person, and clearly establishes a distinct niche, but it completely lacks a 'when to use' trigger clause and offers only two concrete actions with no keyword variations. Adding an explicit trigger phrase and slightly broader action coverage would lift it substantially.

Suggestions

Add an explicit trigger clause, e.g. 'Use when a round evidence packet is ready and a round result must be returned for script import.' — this removes the completeness cap at 3.

Expand the action list beyond the two current actions (verify evidence, return result) to cover the concrete checks performed (e.g., validating integration digests, promotion evidence, cleanup verification) for higher specificity.

Include natural keyword variations such as 'round verification', 'round review', 'round pass/blocked' so users' phrasing matches the description triggers.

Dimension	Reasoning	Score
Specificity	The description states 'Verify integrated round evidence and return a machine-readable round result for script import' — it names the domain and two concrete actions (verify evidence, return a machine-readable result), but coverage is not comprehensive, matching the 1-2-concrete-actions anchor. Not 4 because only two actions are listed rather than 'several specific actions'; not 2 because both actions named are concrete, not generic.	3 / 5
Completeness	The 'what' is clear (verify round evidence, return a machine-readable result), but there is no 'Use when...' clause or any explicit trigger guidance, which caps completeness at 3 per the judging guidelines. It is not 4 because 'when' is entirely absent rather than just imprecise, and not 2 because the 'what' half is concrete.	3 / 5
Trigger Term Quality	Relevant keywords exist ('round evidence', 'round result', 'verify', 'machine-readable', 'script import') but there are no natural trigger phrases a user would say and no synonyms or variations of the core terms. It has some relevant keywords yet misses common variations, matching anchor 3; anchor 4 requires good coverage with only a few natural terms missing.	3 / 5
Distinctiveness Conflict Risk	Terms like 'integrated round evidence', 'round result', and 'script import' occupy a clear niche with distinct triggers and minimal overlap with generic skills. Anti-drift check against anchor 4: the niche is sharper than 'minor overlap risk with closely related skills', so 5 is the best fit.	5 / 5
	Total	14 / 20 Passed

Validation

100%

Warnings & errors only

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: SeemSeam/claude_codex_bridge
Path: docs/plantree/plans/agentic-loop-workflow/drafts/agentroles.ccb_round_reviewer/skills/round-verification/SKILL.md
Commit: 18bf0af

Reviewed: 7 days ago

Table of Contents

Discovery Implementation Validation

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.