CtrlK
BlogDocsLog inGet started
Tessl Logo

gateguard

PreToolUse fact-forcing gate that denies the first Edit/Write/Bash (including MultiEdit) attempt until the agent presents concrete facts (importers, data schemas, verbatim user instruction), then allows retry; A/B-tested at +2.25 quality points. Use when enabling or configuring the GateGuard hook, exempting paths via env vars, or handling first-touch denials.

62

Quality

75%

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 ./skills/gateguard/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

57%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 delivers concrete, well-templated guidance for gate behavior, env var configuration, and parallel-batch handling, but it is padded with marketing prose and redundant disable instructions, lacks a verification step for the setup workflow, and references a bundled script that does not exist in the skill's directory.

Suggestions

Trim the Core Concept/Evidence positioning prose and consolidate the three separate 'turn the gate off' sections into one, keeping the token budget for operational detail.

Add a concrete hooks.json/settings.json snippet showing exactly how to enable `scripts/hooks/gateguard-fact-force.js`, plus a one-step verification (e.g. trigger a test edit and confirm the denial fires).

Actually bundle the referenced hook script (or drop the 'included in this plugin' claim), and consider moving the glob-semantics deep dive into a references/ file linked from SKILL.md.

DimensionReasoningScore

Conciseness

The body is mostly efficient but carries noticeable padding: the motivational Core Concept prose ("LLM self-evaluation doesn't work... This is verified experimentally"), the "No competitor does all three" positioning line, a full Evidence A/B table, and disable instructions repeated across three sections (ECC_GATEGUARD=off appears in Quick Start, Graduated controls, and Turning the gate off). This exceeds the 'minor instances' that anchor 4 tolerates.

3 / 5

Actionability

Gate prompt templates, the env var table with defaults and exact accepted spellings, glob semantics, and the pip install commands are concrete and executable, but Option A's key step — 'Enable it via hooks.json' — provides no actual hooks.json or settings snippet showing how to enable the hook, leaving a minor gap.

4 / 5

Workflow Clarity

The DENY→FORCE→ALLOW sequence and the parallel-batch recovery loop ("present the facts, retry the denied edit, and re-read the file") are clearly sequenced, but the setup workflow has no validation checkpoint — nothing verifies the hook is active or that env var configuration took effect. This matches 'sequence present but checkpoints missing or implicit' better than anchor 4.

3 / 5

Progressive Disclosure

No bundle files exist (no references/, scripts/, or assets/), yet the body claims the hook at `scripts/hooks/gateguard-fact-force.js` is 'included in this plugin' — a reference to a file that is not present. At ~200 lines with inlined glob-semantics detail and full env-var tables, the content could be better split; structure exists but organization and reference signaling are incomplete.

3 / 5

Total

13

/

20

Passed

Description

92%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 strong description: it states the gate mechanism concretely and explicitly, provides a clear 'Use when...' clause, and stays in third person. The only weakness is slightly incomplete trigger-term coverage (missing disable/bypass phrasings), and the embedded "A/B-tested at +2.25 quality points" claim reads more as promotion than trigger guidance.

Suggestions

Add natural trigger phrasings users would say, e.g. 'Use when turning off or disabling GateGuard, or when the gate keeps blocking edits'.

Drop or shorten the '+2.25 quality points' evidence claim from the description; it consumes trigger space without helping a user find the skill.

Mention the pip package name ('gateguard-ai') in the description since users may reference the installable package when asking for help.

DimensionReasoningScore

Specificity

The description lists multiple specific concrete actions — "denies the first Edit/Write/Bash (including MultiEdit) attempt until the agent presents concrete facts (importers, data schemas, verbatim user instruction), then allows retry" — covering the full gate mechanism comprehensively, matching the anchor-5 example's coverage level.

5 / 5

Completeness

It explicitly answers both what ("PreToolUse fact-forcing gate that denies the first Edit/Write/Bash attempt... then allows retry") and when ("Use when enabling or configuring the GateGuard hook, exempting paths via env vars, or handling first-touch denials") with concrete trigger phrases, in third-person voice.

5 / 5

Trigger Term Quality

"enabling or configuring the GateGuard hook, exempting paths via env vars, or handling first-touch denials" gives good natural keyword coverage, but common user phrasings like "turn off/disable GateGuard", "the gate is blocking my edit", or "bypass the gate" are missing, so it does not reach comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

The description carves out a clear niche — a named PreToolUse fact-forcing gate with distinct triggers (hook configuration, path exemption, first-touch denials) — creating minimal conflict risk with other skills.

5 / 5

Total

19

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

referenced_paths_exist

Referenced path issues: 1 missing, 1 deeper-than-1-level

Warning

Total

14

/

16

Passed

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.