CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

agentset-webhooks

Receive and verify Agentset webhooks (managed RAG platform). Use when setting up an Agentset webhook endpoint, verifying the agentset-signature header, debugging 401s, or handling document.ready, document.error and ingest_job.ready events.

Quality

90%

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

SKILL.md
Quality
Evals
Security

Quality

Content

82%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.

An information-dense, highly actionable body built around provider-specific facts and copy-paste verification code, with well-signaled references that all resolve except the examples/ paths. The main deductions are the promotional tail sections (Recommended + Related Skills, with a duplicated entry) and the absence of a single consolidated handler sequence or in-bundle examples to back the 'complete handlers' pointer.

Suggestions

Consolidate the 'Recommended: webhook-handler-patterns' and 'Related Skills' sections: webhook-handler-patterns appears in both with overlapping link lists — a single line per related skill (or one merged section) would reclaim ~20 lines of pure promotion.

Add a short numbered end-to-end sequence (read raw body → verify signature → 401 on fail → parse → dispatch on top-level event → respond 2xx within 20s → dedupe on envelope id → defer slow work) so the workflow reads as one coherent order instead of being assembled across three sections.

Fix or fulfill the examples/ references: the body cites examples/express/, examples/nextjs/ and examples/fastapi/ for 'complete handlers with tests' but no examples/ directory exists in the bundle — either ship it or remove/redirect those links (e.g., to the repository URL).

DimensionReasoningScore

Conciseness

The core sections are dense and every token earns its place — provider facts Claude cannot know (raw-body HMAC, no timestamp signed, retry schedule '12 s, 2m24s, 30m8s', 20-second 2xx deadline, 20th failure disables) with zero padding or explanation of known concepts. Sits at 4 rather than 5 because ~20 lines of tail content could be trimmed: the 'Recommended' section's four external link bullets and the nine-bullet 'Related Skills' list are cross-promotion, and webhook-handler-patterns is listed in both sections.

4 / 5

Actionability

Copy-paste-ready executable verification code with input validation, timing-safe comparison, and the express.raw reading instruction; plus concrete commands ('npx hookdeck-cli listen 3000 agentset --path /webhooks/agentset'), an env var with its dashboard location, explicit 401-on-mismatch failure behavior, and a Python pointer ('hmac.compare_digest on bytes'). Fully executable and covers the common cases, matching the top anchor.

5 / 5

Workflow Clarity

The sequence is explicit and in the right order — read the body raw, verify first ('parse only after the check'), 401 on malformed/mismatch, dispatch on the top-level event, respond 2xx within 20 seconds, 'Make handlers idempotent (dedupe on the envelope id) and do slow work asynchronously' — with the signature check acting as a built-in validation checkpoint. Not 5 because no single end-to-end ordered sequence is presented in SKILL.md: the steps are distributed across the Verification, Event Types, and Delivery sections, and the complete handler flow is deferred to the examples/ directory.

4 / 5

Progressive Disclosure

Good structure with clearly signaled one-level-deep references: the three references/*.md files listed in Resources each exist in the bundle with a one-line description, and the event table/env-var/retry content is appropriately inlined. Not 5 because the body repeatedly points to examples/express/, examples/nextjs/, examples/fastapi/ and examples/ ('For complete handlers with tests, see...') yet no examples/ directory exists in the bundle — dangling key references are a navigation gap beyond the 'minor organization gaps' of the top anchor.

4 / 5

Total

17

/

20

Passed

Description

95%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 model third-person description: a concise what-clause paired with an explicit 'Use when' trigger list naming the vendor-specific header, failure mode (401s), and exact event names. Only minor compression of the what-clause keeps specificity from the top anchor.

DimensionReasoningScore

Specificity

The 'what' names two concrete actions ('Receive and verify Agentset webhooks') and the trigger clause enumerates specific operations ('setting up an Agentset webhook endpoint, verifying the agentset-signature header, debugging 401s, or handling document.ready, document.error and ingest_job.ready events'). This matches the 'several specific actions; minor gaps' anchor — not 5 because the what-clause itself covers only receive/verify, omitting other capabilities the skill supports (retries, local-development tunneling).

4 / 5

Completeness

Explicitly answers both questions with concrete trigger phrases: 'Receive and verify Agentset webhooks (managed RAG platform)' states what, and 'Use when setting up an Agentset webhook endpoint, verifying the agentset-signature header, debugging 401s, or handling... events' states when. This mirrors the rubric's top anchor example almost exactly; the 4 anchor's 'when could be more explicit' caveat does not apply.

5 / 5

Trigger Term Quality

Covers the natural terms a user would actually say or paste: 'Agentset webhook endpoint', 'agentset-signature header', 'debugging 401s', and the exact event names 'document.ready', 'document.error', 'ingest_job.ready'. These are the comprehensive natural tokens of this domain (no file extensions apply), matching the top anchor rather than the 'a few natural terms missing' anchor at 4.

5 / 5

Distinctiveness Conflict Risk

The Agentset brand plus provider-specific tokens ('agentset-signature', 'document.ready', 'ingest_job.ready') give it a clear niche with minimal overlap risk even against sibling webhook-verification skills. It is not the 4 anchor, since the trigger terms are vendor-unique rather than merely 'mostly distinct'.

5 / 5

Total

19

/

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

relative_links

Relative link issues: 4 missing

Warning

Total

15

/

16

Passed

Repository

hookdeck/webhook-skills

Path

skills/agentset-webhooks/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.