CtrlK
BlogDocsLog inGet started
Tessl Logo

strategic-compact

任意の自動コンパクションではなく、タスクフェーズを通じてコンテキストを保持するための論理的な間隔での手動コンパクションを提案します。

56

Quality

64%

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 ./docs/ja-JP/skills/strategic-compact/SKILL.md
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 lean, well-structured skill document: it explains the strategy briefly, gives copy-paste hook configuration with concrete thresholds, cleanly separates plugin and manual installation, and distills best practices without padding. The main gaps are the absence of the referenced script/installer in the bundle and the lack of any post-installation verification guidance.

Suggestions

Bundle 'suggest-compact.js' (and 'install.sh') under scripts/ so the skill is self-contained, or state clearly where they are obtained.

Add a short verification step after hook setup (e.g. run a test Edit and confirm the suggestion appears) to close the workflow's feedback loop.

Trim the 'なぜ' section's restatement of auto-compaction's drawbacks and replace the bare external link with a properly labeled reference document.

DimensionReasoningScore

Conciseness

The body is efficient — each section carries new information (strategic boundaries, script behavior with concrete defaults, hook JSON, env var, best practices) with minimal padding. The 'なぜ戦略的コンパクションか?' rationale section re-states the description's contrast and explains auto-compaction behavior Claude already knows, which keeps it below anchor 5; it is well above anchor 3 because there is no substantial over-explanation.

4 / 5

Actionability

The hook configuration JSON is copy-paste ready, thresholds and defaults are explicit ('COMPACT_THRESHOLD', default 50, reminders every 25), and the plugin-vs-manual install paths are concretely distinguished with an anti-duplication warning. It falls short of anchor 5 because the core script 'suggest-compact.js' and 'install.sh' are referenced but not bundled or shown, leaving a gap in what a user can actually execute from this file alone.

4 / 5

Workflow Clarity

The mechanism is sequenced in three numbered steps (track tool calls, threshold detection, periodic reminders) and the setup path is unambiguous for both install modes, including an explicit warning against double registration. It is a simple config skill with no destructive or batch operation, so no validation checkpoint is required; it misses anchor 5 only because there is no guidance for verifying the hook actually fires after installation.

4 / 5

Progressive Disclosure

The body is a well-organized single file (~60 lines) with clear sections and appropriately short for a standalone SKILL.md; no bundle files exist, so nothing is improperly inlined or deeply nested. It is below anchor 5 because the referenced script and installer are not present in the bundle and the only external pointer is a bare URL to a social-media post rather than a clearly signaled reference document.

4 / 5

Total

16

/

20

Passed

Description

53%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 clearly and concretely states what the skill does in third person, but it omits any 'when to use it' guidance and lacks the natural trigger phrases ('compact', 'context window', '/compact') that would let a user or model select it reliably. It is a serviceable one-sentence summary rather than a complete, trigger-rich description.

Suggestions

Append an explicit trigger clause, e.g. 'Use when the user mentions compaction, compacting the context, /compact, running out of context window, or long multi-phase tasks.'

Add the concrete mechanisms to the 'what' (tracking tool calls, suggesting /compact at a configurable threshold) so the capability list is more comprehensive.

Include natural synonyms and command forms — '/compact', 'context window', 'token budget' — so the description matches how users actually phrase the need.

DimensionReasoningScore

Specificity

The description names the domain ('compaction') and one concrete action ('proposes manual compaction at logical intervals to preserve context through task phases'), but lists no further capabilities or configuration hooks. It matches anchor 3 (domain plus 1-2 concrete actions, not comprehensive); it is below anchor 4 because no several specific actions are enumerated, and above anchor 2 because the action stated is concrete, not generic.

3 / 5

Completeness

The 'what' is clearly stated ('proposes manual compaction at logical intervals... rather than arbitrary auto-compaction'), but there is no 'Use when...' clause or equivalent trigger guidance, which per the rubric caps completeness at 3. It is not anchor 4 because 'when' is entirely absent rather than weakly implied, and not anchor 2 because the 'what' is concrete, not vague.

3 / 5

Trigger Term Quality

Relevant domain keywords are present ('自動コンパクション', '手動コンパクション', 'コンテキスト', 'タスクフェーズ') but common user-facing variations are missing — no '/compact', 'context window', 'token limit', or synonyms. Anchor 3 (some relevant keywords, missing common variations) fits best: above anchor 2 because the terms are domain-specific rather than generic, below anchor 4 because key natural trigger phrases are absent.

3 / 5

Distinctiveness Conflict Risk

Compaction-strategy is a narrow niche with few overlapping skills, and the description's contrast with arbitrary auto-compaction distinguishes it further. Anchor 4 (mostly distinct, minor overlap risk with closely related skills) fits: not anchor 5 because no explicit trigger phrases establish the niche, and not anchor 3 because the domain is specific rather than merely 'somewhat specific'.

4 / 5

Total

13

/

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
affaan-m/ECC
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.