CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

cross-thread-sync

跨 thread 协同:通知、归属核验、争用与责任处置。Use when: 平行 session 通知或共享文件争用。Not for: 跨猫交接或新建 thread。Output: routed cross-post + disposition。GOTCHA: ACTION/BLOCKING 不转移球权。

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

70%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 delivers a well-sequenced, validation-heavy workflow with concrete tool invocations and rich error-recovery guidance — genuinely actionable for its domain. It is held back by a monolithic single-file structure with several sections that belong in reference files, and by noticeable repetition of the routing rule across three sections.

Suggestions

State the @mention/targetCats routing rule once (e.g. in the 通知 step) and have the misconception table reference it instead of restating it, freeing budget in a ~240-line file.

Split self-contained policy sections (爪感差/F245 special routing, F246 Phase J dispatch lifecycle, Ghost Thread Bug rules) into one-level-deep reference files under references/ and link them from a short overview.

Tighten the [BLOCKING] table cell by moving the S.1-c recovery-sweep policy into the 处置 section as prose, keeping the table scannable.

DimensionReasoningScore

Conciseness

Almost all content is novel domain policy Claude could not infer (routing rules, claim protocol, custody semantics) rather than padding, but the critical routing rule (@mention/targetCats) is restated three times (top blockquote, Step 2 warning, misconception table), and the [BLOCKING] table cell and 常见误区 rows embed long overlapping policy paragraphs that could be consolidated. Fits 'mostly efficient but could be tightened'; not 4 because the redundancy goes beyond minor instances.

3 / 5

Actionability

Concrete tool calls with parameters (cat_cafe_feat_index(featId="F088"), cat_cafe_list_threads(activeSince=...), a full cat_cafe_cross_post_message(...) invocation with content template) plus scenario-to-action tables and an explicit claim/release protocol with coordination phases. Not 5: claim/release are shown as quoted message text rather than complete tool invocations, and some cells remain policy-level hints (e.g. "带 Decision Packet @co-creator") without executable form.

4 / 5

Workflow Clarity

A clear four-step pipeline (发现 → 通知 → 协调 → 处置) with an explicit inbound validation gate ("三问(缺一不接)"), a judgment-to-action table with verified/mismatch/insufficient verdicts, a failure-mode callout, and a claim protocol with timeout/re-claim recovery and operator escalation — explicit checkpoints plus feedback loops, matching the top anchor.

5 / 5

Progressive Disclosure

No bundle files exist and everything lives in one ~240-line monolith; self-contained policy sections (爪感差/F245 routing, F246 Phase J lifecycle, Ghost Thread Bug rules) would naturally be separate reference files, and the only external pointer (docs/bug-report/...) is a repo path rather than a skill reference. Section headers do give reasonable structure, so this fits 'some structure; content that should be separate is inline' rather than the minimal-structure anchor at 2 or the well-split anchor at 4.

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 dense and highly informative: concrete actions, an explicit Use when clause, explicit Not-for boundaries against the sibling handoff skill, a stated output, and a known-gotcha warning. Its main weakness is jargon-heavy terseness (metaphorical terms like 球权) and missing plain-language synonyms in the trigger phrasing.

DimensionReasoningScore

Specificity

"跨 thread 协同:通知、归属核验、争用与责任处置" names the domain plus several concrete actions (notification, ownership verification, contention handling, custody disposition), and "Output: routed cross-post + disposition" specifies the deliverable. Not 5: the action list is terse and relies on internal jargon (e.g. "球权"), leaving minor coverage gaps; not 3: it lists several distinct actions, more than 1-2.

4 / 5

Completeness

Explicitly answers what ("通知、归属核验、争用与责任处置"), when ("Use when: 平行 session 通知或共享文件争用"), plus boundaries ("Not for: 跨猫交接或新建 thread") and expected output. Both what and when are concrete and explicit, matching the top anchor; a 4 would require the when-clause to be less specific, which it is not.

5 / 5

Trigger Term Quality

"平行 session 通知或共享文件争用" and "跨 thread 协同" are natural phrases a user would say for this need. Not 5: the description itself omits common synonyms and plain-English variants (e.g. "cross-thread", "notify another session"); not 3: coverage is good, not just partial.

4 / 5

Distinctiveness Conflict Risk

"Not for: 跨猫交接或新建 thread" explicitly carves out the adjacent cross-cat-handoff skill and thread creation, giving a clear niche with minimal conflict risk. It is more distinct than the 'minor overlap' anchor at 4.

5 / 5

Total

18

/

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

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

zts212653/clowder-ai

Path

cat-cafe-skills/cross-thread-sync/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.